Three-Dimensional Dynamic Simulation and Infrared Measurement of the GTAW Temperature Field

Literature Overview and Research Context

The paper by Lei Yucheng and colleagues from Jiangsu University, published in 2008 in the Journal of Jiangsu University (Natural Science Edition), addresses a fundamental challenge in welding engineering: the accurate characterization of temperature fields during the rapid heating and cooling cycles of gas tungsten arc welding (GTAW). The correct description of the weld temperature field is the prerequisite for subsequent analysis of microstructural transformations and mechanical properties of welded joints. This work is particularly significant because temperature field accuracy directly influences predictions of residual stress, distortion, and metallurgical evolution in the heat-affected zone (HAZ).

The research was supported by the National Natural Science Foundation of China (Grant No. 50475126), reflecting the recognition that thermal modeling is a critical enabling technology for welding process optimization. The authors developed a three-dimensional dynamic finite element model of the welding temperature field under arc movement, utilizing the commercial finite element analysis software ANSYS with a custom APDL (ANSYS Parametric Design Language) program.

Core Technical Methodology

The computational approach employed a double elliptical heat source model, which is widely regarded as one of the most effective representations for arc welding heat input. This model accounts for the asymmetric heat distribution between the front and rear of the moving arc, with the front ellipse representing the rapid heating zone and the rear ellipse representing the slower cooling zone. The double elliptical model was selected over simpler alternatives such as the Gaussian point source or the single elliptical model because it better captures the physical reality of the GTAW process, particularly the deeper penetration on the leading edge.

A critical aspect of this study is the incorporation of temperature-dependent thermophysical properties of the aluminum alloy material. The authors explicitly considered three major heat transfer mechanisms: radiation, convection, and phase transformation. Each of these contributes significantly to the accuracy of the thermal simulation:

The dynamic simulation was conducted by sequentially activating heat source elements as the arc moved, which is computationally efficient compared to full 3D transient analysis of the entire domain at once. This moving heat source approach is standard practice in welding thermal modeling and was implemented through the APDL program within ANSYS.

Experimental Validation via Infrared Thermometry

The experimental validation employed a SAT-HY6800 infrared temperature measurement instrument to capture the temperature distribution in the weld zone during actual GTAW welding of aluminum alloy. A significant technical challenge in infrared measurement of welding processes is the overwhelming radiation from the electric arc itself, which can saturate the detector and produce meaningless readings. The authors addressed this by installing a filter that selectively blocks the arc radiation while allowing the thermal infrared emission from the workpiece surface to pass through. This is a practical and important engineering solution, as many practitioners encounter this problem when attempting non-contact temperature measurement during arc welding.

The comparison between measured and calculated temperatures showed good agreement, validating both the finite element model and the heat source parameters used. This level of validation is essential before using such models for predictive purposes in engineering practice, such as optimizing welding parameters or predicting HAZ microstructure.

Engineering Practice Implications and Reflections

From a practical standpoint, this work demonstrates the value of coupled computational and experimental approaches in welding research. In my experience with steel pipe manufacturing and fitting fabrication, temperature field predictions are invaluable for several reasons. First, they allow pre-qualification of welding procedures without extensive trial welds, which is particularly important for expensive materials such as corrosion-resistant alloy (CRA) pipes or thick-section butt-weld fittings. Second, accurate thermal models enable prediction of cooling rates, which directly determine grain structure in the HAZ and, consequently, mechanical properties.

For aluminum alloy welding specifically, the rapid thermal cycling and high thermal conductivity present unique challenges. The thermal model developed here can be adapted for other materials by substituting appropriate thermophysical property functions. However, practitioners should be aware that the double elliptical model, while widely used, has limitations: it assumes a constant heat input distribution shape regardless of material, and it does not account for the influence of weld pool dynamics on heat distribution. More advanced models using coupled electromagnetic-fluid-thermal approaches exist but come at significant computational cost.

The infrared measurement technique with arc radiation filtering is directly applicable to production environments. In my work with pipe fitting fabrication, I have seen similar approaches used to monitor welding quality in real time. The key lesson is that non-contact measurement, when properly configured, provides valuable process monitoring capability without interfering with the welding operation.

This study represents a solid contribution to welding thermal modeling methodology, and its validation approach provides a template for engineers who wish to implement similar analyses for their own welding processes.