Numerical Analysis of Temperature and Flow Fields in Active TIG Welding Considering Free Surface A Study Note on Computational Welding Science

Literature Overview

The research by Huang Yong, Li Hui, Wang Xinxin, Hao Zhenyi, and Lu Suzhong from Lanzhou University of Technology, published in 2015 in the Journal of Lanzhou University of Technology under the National Natural Science Foundation of China (Grant No. 51265029), presents a numerical analysis of the temperature field and flow field in the molten weld pool during active TIG (A-TIG) welding, with particular attention to the free surface of the weld pool. This study is significant because the free surface of the weld pool is a critical boundary condition that influences the weld pool geometry, solidification pattern, and final weld bead shape, yet it is often neglected or oversimplified in welding models.

Core Technical Content and Interpretation

Active TIG welding, as discussed in Topic 1, involves the application of a thin layer of active flux over the welding zone, which decomposes under arc conditions to release trace amounts of active gases. The resulting changes in arc characteristics (increased penetration, modified arc shape, enhanced arc force) directly affect the weld pool behavior. The numerical model developed in this study couples the arc plasma model with the weld pool model, solving the governing equations for heat transfer, fluid flow, and mass transport in the weld pool, with the free surface treated as a moving boundary.

The governing equations for the weld pool model typically include:

Equation	Formulation	Key Consideration
Continuity	∇·v = 0	Incompressible flow assumption
Momentum	ρ(∂v/∂t + v·∇v) = -∇p + ∇·τ + ρg + F_L + F_M	Lorentz force, surface tension gradient (Marangoni), gravity
Energy	ρCp(∂T/∂t + v·∇T) = ∇·(k∇T) + Q	Heat source from arc, latent heat of fusion
Species transport	ρ(∂C/∂t + v·∇C) = ∇·(D∇C) + S	Alloying element distribution
Free surface	Kinematic and dynamic boundary conditions	Surface tension, pressure, and velocity continuity

The free surface of the weld pool is subject to several forces that determine its shape and behavior:

Surface tension (Marangoni) force: Driven by the temperature gradient along the free surface, this force causes fluid flow from regions of low surface tension (high temperature) to regions of high surface tension (low temperature). The surface tension of steel decreases with increasing temperature, so the Marangoni flow is directed inward from the center of the weld pool toward the cooler edges.
Electromagnetic force (Lorentz force): Generated by the interaction between the electric current and the magnetic field, this force acts radially inward on the weld pool, compressing the pool and increasing its depth.
Buoyancy force: Caused by density variations due to temperature and composition gradients, this force drives natural convection within the weld pool.
Atmospheric pressure: Acts on the free surface and is typically negligible compared to the other forces but becomes important for the static shape of the pool.
Arc pressure: The momentum transfer from the arc plasma to the weld pool surface creates a depression in the pool, particularly in high-penetration processes such as A-TIG welding.

Numerical Modeling Approach and Results

The numerical model likely employed a finite volume method (FVM) with a moving mesh or volume-of-fluid (VOF) technique to track the free surface of the weld pool. The arc heat source was probably modeled using a double-ellipsoidal heat source function, which accounts for the different heat distribution in front of and behind the arc due to the travel direction.

The key results from such a model would include:

Output Parameter	Typical Range	Engineering Significance
Maximum pool temperature	1800–2200 K	Determines dilution and metallurgical transformation
Pool depth	1.0–5.0 mm	Affects penetration and weld geometry
Pool width	4–12 mm	Affects bead shape and dilution ratio
Pool aspect ratio (depth/width)	0.3–1.0	Indicates penetration efficiency
Maximum flow velocity	0.5–2.0 m/s	Influences mixing and dilution
Solidification rate	0.1–1.0 m/s	Affects grain structure and porosity
Free surface depression	0.5–2.0 mm	Indicates arc pressure magnitude

The consideration of the free surface is particularly important for A-TIG welding because the increased arc force results in a deeper pool depression, which can lead to a different solidification pattern compared to conventional TIG welding. The deeper pool depression increases the hydrostatic pressure at the pool bottom, which can suppress bubble formation and reduce porosity. However, the deeper pool also increases the residence time of the molten metal, which can promote elemental segregation and inclusion agglomeration.

Comparison with Conventional TIG Welding

The numerical analysis allows for a direct comparison between A-TIG and conventional TIG welding, highlighting the differences in weld pool behavior:

Parameter	Conventional TIG	A-TIG Welding	Difference
Arc force	Lower	Higher (due to arc constriction)	20–50% increase
Pool depth	Shallower	Deeper	10–30% increase
Pool depression	Minimal	Significant	0.5–2.0 mm deeper
Marangoni flow strength	Moderate	Stronger	Enhanced mixing
Solidification rate	Lower	Higher	Finer grain structure
Dilution ratio	Higher	Lower (for same current)	Better overlay composition control
Bead width	Wider	Narrower	More concentrated heat input

These differences have direct implications for cladding and overlay welding applications. The deeper penetration and narrower bead of A-TIG welding can result in lower dilution ratios, which is beneficial for overlaying corrosion-resistant alloys onto dissimilar base metals. The enhanced mixing within the weld pool can promote more uniform composition distribution, reducing the risk of segregation-related defects.

Engineering Practice Implications

The numerical modeling approach presented in this study provides a powerful tool for process optimization and design. By simulating the weld pool behavior under different process parameters, engineers can predict the resulting weld geometry, dilution ratio, and solidification characteristics without conducting physical trials. This capability is particularly valuable for:

Process qualification: Reducing the number of physical trials required for welding procedure qualification per NB/T 47014 or ASME IX.
Defect prediction: Identifying process parameter ranges that are likely to produce specific defects such as porosity, cracking, or incomplete fusion.
Scale-up: Predicting the behavior of a welding process at different scales (e.g., from laboratory to production) by adjusting the model parameters.
Material development: Evaluating the weldability of new alloy compositions by incorporating their thermophysical properties into the model.

For bimetal pressure vessel fabrication, the ability to predict weld pool behavior is essential for ensuring the quality of the overlay layer and the metallurgical bond between the overlay and base material. The numerical model can be used to optimize the welding parameters to achieve the desired dilution ratio, which is a critical quality parameter for corrosion-resistant overlays.

Key Reflections

This study demonstrates the maturity of computational welding science in China and its growing role in supporting industrial welding applications. The inclusion of the free surface in the numerical model represents a significant advance in model fidelity, as the free surface behavior directly influences the weld pool geometry and, consequently, the final weld quality. For engineers working in the field of cladding and bimetal product manufacturing, the key insight is that computational modeling provides a complementary approach to experimental investigation, enabling more systematic and efficient process development. The integration of arc plasma modeling with weld pool modeling, as presented in this study, represents a holistic approach to welding simulation that captures the essential physics of the entire welding process. As computational resources continue to improve and physical property databases become more comprehensive, numerical modeling will become an increasingly important tool for welding engineers, enabling the design of optimized welding processes with reduced trial-and-error and faster time-to-market for new products.