A comparative study of power and vibration signals for different tool pin sizes of friction stir welding

Abstract

Process signals play an important role in revealing weld formation, reflecting the interaction between mechanical mixing and process parameters. This study focuses on how different tool pin lengths and material thicknesses influence the signals generated during friction stir welding of AA6061. Three pin lengths that were 2 mm, 4 mm, and 6 mm were tested on plates with thicknesses of 4 mm, 6 mm, and 8 mm (made by stacking two 4 mm plates). Spindle speed of 1200 RPM and a traverse speed of 50 mm/min were constant. Shorter pins cannot fully penetrate thicker plates in a single pass; therefore, welding was performed in dual passes, top and bottom, to complete the joint. During welding, power consumption and vibration signals were recorded. Data collected from the bottom side were used for detailed analysis, as the top pass helped hold materials before full penetration. The signals were analyzed using the Fast Fourier Transform, and standard deviation and defect accuracy were calculated to evaluate the weld condition. The findings show that the 2 mm pin achieved the highest defect accuracy of 90.63% based on power signals, while the 6 mm pin achieved the highest accuracy of 85.10% based on vibration signals. Larger vibration standard deviations from the 6 mm pin indicate energetic material stirring. Hardness results suggest that the 4 mm pin provides the most favorable material consolidation. These outcomes highlight that pin length and plate thickness significantly affect process signals, making them crucial variables for real-time weld quality monitoring and optimization.