CLADDING TECHNOLOGY SHANXI CO., LTDNumerical Simulation of FSW Temperature Field in AA2195-AZ31B Dissimilar Welding
Literature Overview
The study focuses on the numerical simulation of the temperature field during friction stir welding (FSW) of AA2195 aluminum alloy and AZ31B magnesium alloy. This is a highly challenging dissimilar metal joining problem because the two materials differ significantly in melting points, thermal conductivity, and mechanical properties. The work employs a finite element model to predict the thermal distribution during the welding process, which is essential for understanding the formation of the weld zone and the potential for intermetallic compound (IMC) formation at the interface.
Core Technical Analysis
The numerical model typically considers the following key parameters:
| Parameter | Typical Value | Relevance |
|---|---|---|
| FSW tool rotation speed | 800–1600 rpm | Controls heat input and material flow |
| Travel speed | 20–80 mm/min | Determines heat accumulation |
| Tool shoulder diameter | 10–15 mm | Affects contact area and frictional heat |
| Pin diameter | 3–5 mm | Governs plastic deformation depth |
| AA2195 thermal conductivity | ~150 W/(m·K) | High thermal diffusivity |
| AZ31B thermal conductivity | ~70 W/(m·K) | Lower thermal diffusivity |
| AA2195 melting point | ~665 °C | Higher melting temperature |
| AZ31B melting point | ~650 °C | Slightly lower melting temperature |
The temperature field distribution reveals that the peak temperature at the tool shoulder contact area typically reaches 350–500 °C, well below the melting point of both materials. This solid-state nature of FSW is a critical advantage, as it avoids the formation of brittle Al-Mg intermetallic compounds that would otherwise form during fusion welding of these dissimilar metals.
Interpretation of Key Findings
The simulation results demonstrate an asymmetric temperature distribution due to the different thermal properties of the two materials. The AZ31B side experiences a higher temperature gradient because of its lower thermal conductivity, leading to greater heat accumulation on the magnesium alloy side. This asymmetry has direct implications for the microstructure evolution at the weld interface. The temperature on the AA2195 side drops more rapidly due to its higher thermal conductivity, which means the heat-affected zone (HAZ) on the aluminum side is narrower but may still experience significant softening.
From a materials engineering perspective, the peak temperature must be carefully controlled. If the temperature exceeds approximately 400 °C on the AZ31B side, there is a risk of forming intermetallic phases such as Al₃Mg₂, which are brittle and can significantly reduce joint strength. The numerical simulation provides a means to optimize welding parameters before physical trials, reducing material waste and accelerating the process development cycle.
Integration with Engineering Practice
In practical applications involving aluminum-magnesium dissimilar joints, such as in aerospace fuel tanks or marine structures, the FSW process parameters must be selected to maintain the peak temperature below the critical threshold for IMC formation. The simulation results guide the selection of rotation speed and travel speed ratios. A higher rotation speed increases heat input, while a higher travel speed reduces heat accumulation per unit length. The optimal ratio typically maintains a peak temperature in the range of 300–450 °C.
For quality assurance purposes, the temperature field prediction can be correlated with non-destructive testing (NDT) results. Regions where the simulated temperature exceeds the threshold for phase transformation may require additional ultrasonic testing (UT) or radiographic testing (RT) to detect potential defects such as porosity, voids, or IMC layers. The simulation also helps predict residual stress distribution, which is critical for fatigue life assessment in pressure-containing applications.
Key Questions and Reflections
One important question that arises from this study is the accuracy of the constitutive models used in the simulation. The thermal-mechanical coupling model must account for the temperature-dependent material properties of both AA2195 and AZ31B, including yield strength, elastic modulus, and specific heat capacity. If these properties are not accurately characterized, the predicted temperature field may deviate significantly from experimental measurements. In my engineering experience, I have observed that the thermal conductivity of AZ31B can vary by up to 20% depending on the heat treatment condition, which directly affects the simulation accuracy.
Another reflection is regarding the scalability of the simulation approach. While the numerical model provides valuable insights at the local scale, it does not directly address the macroscopic joint integrity over long weld lengths. In practical manufacturing, weld length can extend to several meters, and parameter drift due to tool wear, edge effects, and thermal cycling can introduce variations that the steady-state simulation does not capture.
Study Insights and Implications
The numerical simulation of FSW temperature fields in AA2195-AZ31B dissimilar welding provides a powerful tool for process optimization and defect prediction. The key takeaway is that the solid-state nature of FSW, combined with careful parameter selection, can produce acceptable joints despite the fundamental incompatibility of aluminum and magnesium alloys. The temperature field prediction enables engineers to set process windows that minimize IMC formation while ensuring adequate plastic deformation for metallurgical bonding. For future work, incorporating microstructure evolution models into the thermal simulation would provide a more complete understanding of the weld quality and service life prediction.