Molten Pool Free Surface Evolution in Stationary TIG Welding

Literature Overview and Research Motivation

The study by Fan Ding, Huang Lin, Huang Jiankang, and Shi Yu from Lanzhou University of Technology, published in the Journal of Lanzhou University of Technology in 2014 (Vol. 40, No. 5, pp. 14-18), presents a three-dimensional numerical analysis of molten pool free surface evolution during stationary TIG welding. Funded by the National Natural Science Foundation (Grant 51205179), this research employs computational fluid dynamics to model the complex fluid flow and heat transfer phenomena occurring within the weld pool.

Stationary TIG welding, also known as point welding or non-traveling arc welding, is used for stud welding, tack welding, and certain repair applications. Unlike traveling arc welding, the arc remains fixed at a single location, allowing the molten pool to evolve over time under the influence of multiple driving forces. Understanding the free surface evolution is critical for predicting weld geometry, assessing porosity formation risk, and optimizing process parameters.

Numerical Model Development

The researchers developed a three-dimensional model based on fundamental fluid mechanics principles, incorporating several key physical phenomena. The model uses a Gaussian heat source distribution to represent the arc heat input, which provides a realistic representation of the spatial and temporal heat flux distribution.

The Volume of Fluid (VOF) method is employed to track the molten pool free surface. This method solves an additional transport equation for a volume fraction function that distinguishes between liquid metal and gas phases. The VOF approach allows for accurate capture of the free surface shape and its evolution over time without requiring mesh deformation.

The model accounts for four primary driving forces acting on the molten pool:

The electromagnetic force is generated by the interaction between the induced current density and the magnetic field within the molten pool. This force creates a Lorentz force that drives fluid motion, particularly contributing to pool penetration. The Marangoni force, driven by surface tension gradients caused by temperature variations, promotes outward flow along the pool surface.

The model also incorporates the effect of latent heat during solidification and melting, which is essential for accurately predicting the solid-liquid interface movement. The FLOW3D software was used for numerical solution, with custom development to implement the specific physics of TIG welding.

Key Results on Free Surface Evolution

The numerical results reveal that the TIG welding molten pool undergoes significant free surface deformation over time. As the welding duration increases, both penetration depth and pool width increase, forming a characteristically wide and shallow pool geometry.

The pool evolution can be divided into several stages. Initially, the pool is deep and narrow as the arc pressure dominates the fluid flow. As the pool grows, the Marangoni force becomes more significant, promoting outward surface flow and pool spreading. The buoyancy force drives hot metal upward, contributing to surface deformation.

The free surface develops pronounced undulations and deformations as time progresses. These surface irregularities are caused by the interplay between the outward Marangoni flow and the inward electromagnetic stirring. The resulting complex flow pattern creates localized regions of upward and downward flow at the surface, producing the observed undulations.

The formation of a wide and shallow pool morphology is consistent with experimental observations of TIG welds. The wide pool is attributed to the strong outward Marangoni flow along the surface, which spreads the molten metal laterally. The shallow depth results from the balance between downward arc pressure and upward buoyancy and Marangoni forces.

Engineering Implications and Process Optimization

The understanding of free surface evolution has direct implications for weld quality and defect prevention. The surface undulations can lead to porosity formation if gas inclusions become trapped in the solidifying metal. The wide and shallow pool geometry may result in incomplete root fusion in multi-pass welding if the pool does not achieve adequate penetration.

The numerical model provides a tool for predicting weld pool geometry under various process conditions. By adjusting parameters such as current, arc pressure, and heat flux distribution, engineers can optimize the pool shape for specific applications. For example, increasing current increases both pool width and depth, while adjusting the arc pressure can control penetration.

The study also highlights the importance of transient effects in welding. The pool geometry continues to evolve during the welding process, meaning that weld properties may vary along the weld length even with constant parameters. This transient behavior must be considered in welding procedure development and quality assessment.

Study Insights and Methodological Contributions

This research demonstrates the value of computational fluid dynamics in understanding welding phenomena that are difficult to observe experimentally. The VOF-based approach provides a robust method for tracking free surfaces in complex flow fields, and the inclusion of multiple driving forces yields realistic predictions of pool behavior.

The model can be extended to traveling arc welding by incorporating the arc movement, providing insights into weld pool dynamics along the entire weld length. It can also be used to study the effects of different materials, where variations in surface tension coefficients, electrical conductivity, and thermal properties significantly influence pool behavior.

For engineering practice, the key takeaway is that TIG welding pool geometry is the result of a complex balance of forces, and optimizing weld quality requires understanding and controlling these forces through appropriate parameter selection. The numerical approach presented here offers a powerful tool for process development and optimization, complementing experimental methods and empirical knowledge.