Surface roughness characterization of additive manufactured Ti-6Al-4 V based on laser ultrasonic signal

Abstract

Additive manufacturing (AM), especially AM for metals, has become the most promising processing technique in recent years and is widely applied across several industries. Laser ultrasonic testing (LUT) has appeared as a potential method for identifying defects in additive manufactured components, but the rough surface of the additive manufactured component affects the ultrasonic signal. Analyzing the influence of surface roughness on ultrasonic signals and characterizing surface roughness through ultrasonic signals can improve the accuracy of defect detection and monitor the working conditions in the processing, which is a great significance for enhancing the quality of additive manufactured products. In this study, the influence of additive manufactured rough surface with different Ra on the ultrasonic signal has been analyzed. A novel simulation approach for the rough surface was provided, and the created surfaces with different Ra were applied to modeling. The results indicate that when the surface roughness increases, the ultrasonic signals suffer strong diffuse reflection, and the signal energy decreases. When the Ra is small, the surface of the specimen is close to smooth, the reflection of energy is enhanced, and part of the signals are reflected, so the signal energy is slightly lower. The ultrasonic signal of all models has been denoised and reconstructed based on the variational mode decomposition (VMD), and the analytical expression between the reconstructed signal energy and Ra was derived. The laser ultrasonic experiments were further designed for exciting signals from different Ra specimens, and the analytical expression relating reconstructed signals to Ra was also obtained, then the surface roughness of two specimens with unknown Ra was calculated by ultrasonic signals. A new method to obtain the roughness Ra of the additive manufactured components by ultrasonic signal is developed.