CLADDING TECHNOLOGY SHANXI CO., LTDNumerical Simulation of TIG Welding Fluid Flow Heat Transfer and Interface Tracking with Dynamic Mesh
Literature Overview
This study by Li Linmin, Li Baokuan, Liu Lichao, and Cao Xia from Northeast University and Hohai University presents a comprehensive numerical simulation framework for TIG welding that couples fluid flow modeling, heat transfer analysis, and dynamic mesh interface tracking. Funded by the National Natural Science Foundation of China (Grant 51574068), the research addresses one of the most challenging aspects of welding process simulation: accurately capturing the evolving geometry of the weld pool surface during the welding process.
Simulation Methodology
The numerical model employs a three-dimensional moving heat source coupled with fluid flow and solidification analysis. The key innovation lies in the dynamic mesh technique used to track the free surface of the molten weld pool, which changes shape continuously as the arc moves along the weld path.
| Simulation Parameter | Value / Setting |
|---|---|
| Governing equations | Navier-Stokes, energy, mass conservation |
| Heat source model | Double-ellipsoidal (Goldak) |
| Arc power (W) | 1000-2500 |
| Welding speed (mm/s) | 2-8 |
| Mesh elements | 150,000-300,000 |
| Time step (s) | 0.001-0.005 |
| Surface tension model | Temperature-dependent (Bridges) |
| Electromagnetic force | Self-magnetic pressure |
| Marangoni convection | Surface tension gradient |
| Dynamic mesh method | Arbitrary Lagrangian-Eulerian (ALE) |
The double-ellipsoidal heat source model represents the non-uniform energy distribution of the TIG arc, with the front half of the arc depositing energy more intensely than the rear half. This asymmetry is critical for accurately predicting weld pool shape and solidification patterns.
Key Simulation Results
The fluid flow within the weld pool is dominated by two competing mechanisms: electromagnetic (Lorentz) force and thermocapillary (Marangoni) convection. The electromagnetic force drives fluid downward and outward from the arc center, creating a deep penetration pattern. Marangoni convection, driven by the surface tension gradient caused by temperature variation across the pool surface, drives fluid outward at the surface and inward beneath, creating surface flow that influences weld width.
| Weld Pool Parameter | Simulation Result | Experimental Verification |
|---|---|---|
| Maximum pool depth (mm) | 2.5-4.0 | 2.3-3.8 |
| Maximum pool width (mm) | 8.0-12.0 | 7.5-11.5 |
| Pool length (mm) | 6.0-9.0 | 5.5-8.5 |
| Maximum flow velocity (m/s) | 1.5-3.0 | N/A (indirect) |
| Solidification time (s) | 0.5-1.5 | 0.4-1.2 |
| Peak temperature (°C) | 3200-3500 | 3100-3400 |
The dynamic mesh technique successfully captured the evolution of the weld pool surface from initial depression under arc pressure through steady-state flow to final solidification. The simulation predicted a characteristic weld pool shape with a deep central penetration zone and wider surface spread, consistent with macrographic observations of TIG welds.
Heat Transfer Analysis
The temperature field evolution reveals distinct thermal zones within the weld:
- Molten zone: Temperatures above the liquidus (approximately 1500°C for carbon steel, 1380°C for stainless steel), where fluid flow is active and mass transport is significant.
- Solidification zone: Temperatures between solidus and liquidus, where solidification occurs and dendritic microstructures form.
- Heat-affected zone: Temperatures below the solidus but above the recrystallization temperature, where grain growth and phase transformations occur.
- Base metal zone: Temperatures below 600°C, where elastic-plastic deformation occurs but no microstructural change takes place.
The simulation predicted peak cooling rates of 20-50°C/s in the weld center and 5-15°C/s in the HAZ, which are critical parameters for predicting microstructure and mechanical properties. For low-carbon steel, these cooling rates correspond to ferrite-pearlite microstructures in the weld and fine-grained ferrite-pearlite in the HAZ.
Application to Cladding and Bimetal Fabrication
While the study focuses on conventional TIG welding, the numerical framework has direct applicability to cladding and bimetal manufacturing processes:
- Weld overlay cladding: The same fluid flow and heat transfer models can be applied to simulate multi-pass cladding, where the thermal interaction between successive passes affects the final residual stress and microstructure of the cladding layer.
- Dilution prediction: By modeling the fluid flow patterns at the interface between dissimilar metals, the simulation can predict the dilution ratio between base metal and filler metal, which is critical for controlling the composition of clad layers.
- Residual stress prediction: The coupled thermo-mechanical analysis can predict residual stress distributions in clad plates, providing guidance for stress relief procedures.
- Bond strength estimation: The simulation of solidification at the interface between dissimilar metals can provide insights into the formation of intermetallic compounds and their effect on bond strength.
Study Insights and Engineering Value
This numerical study demonstrates that dynamic mesh techniques can effectively capture the complex physics of TIG welding, including the free surface evolution of the weld pool. The accuracy of the simulation results, validated against experimental measurements, provides confidence in using such models for process optimization and design. For cladding engineers, the ability to simulate multi-pass weld overlay with accurate fluid flow and heat transfer predictions represents a powerful tool for reducing trial-and-error experimentation. The model can be extended to predict the effects of process parameters such as arc oscillation, filler wire feeding, and travel speed on cladding layer quality, ultimately enabling rational process design rather than empirical parameter selection. The integration of such simulation capabilities with digital twin concepts in modern manufacturing could significantly accelerate the qualification and optimization of cladding processes for critical pressure vessel applications.