Friction stir welding (FSW) is an advantageous solid-state joining process, suitable for many materials in the energy, aerospace, naval and automotive industries. Like all other welding processes, friction stir welding requires non-destructive evaluation (NDE). The time and resources to preform NDE is expensive. To reduce these costs, nontraditional NDE methods are being developed for FSW. Spectral based defect recognition uses the forces during the welding process to validate weld quality. Although spectral NDE methods have shown promise as an alternative NDE processes, many research welding speeds do not correspond to manufacturing speeds, nor do they explain the relationship between the spectral data and the process. The purpose of this work is to explore the possibility of acquiring additional information about the defect. Namely the defect’s type, location, and magnitude. In this study, welds with “wormhole” defects were produced at 2000, 2500 and 3000 mmpm in 5754 aluminum. The welding process forces and torque were measured and analyzed spectrally. The welded plates were then imaged with x-ray photography, a validated NDE method. It was found that low frequencies (0 – 4 Hz) in the y & z force signals correlate with defect presence in high speed FSW. In addition, the strong correlation between the spectral data and the presence of a defect allowed for defect magnitude predictions. Linear fits were applied to the defect measurements and the spectral data. Large error inhibits the wide use of this prediction method.
College and Department
Ira A. Fulton College of Engineering and Technology
BYU ScholarsArchive Citation
Hunt, Johnathon Bryce, "Defect Detection in Friction Stir Welding by Measureable Signals" (2020). Theses and Dissertations. 8643.
friction stir welding, FSW, nondestructive examination, NDE, defects