Monitoring Resonant-Assisted Deposition (RAD) Process Using Optical Microphones

Additive Manufacturing (AM) has revolutionized modern manufacturing by enabling the production of intricate geometries, customized designs, and highly functional components. Among various AM techniques, ultrasonic-based additive processes, such as Ultrasonic Additive Manufacturing (UAM) and Resonant-Assisted Deposition (RAD), have gained attention due to their ability to achieve solid-state bonding. Unlike fusion-based methods, these techniques eliminate issues like gas porosity, intermetallic formation, and thermal distortions by leveraging ultrasonic energy to create metallurgical bonds between materials. RAD is an emerging ultrasonic-based AM process that enables precise shaping and joining fine metal wires through ultrasonic energy. However, ensuring process stability and defect-free manufacturing in RAD remains challenging due to the complex interactions between ultrasonic vibrations, material properties, and process parameters. Slight deviations in these factors can lead to inconsistencies in bonding, defects in the deposited layers, and reduced structural integrity of the final component. As a result, real-time monitoring methods are essential to assess joint formation, detect defects, and optimize process parameters dynamically. This study investigates the feasibility of using optical microphones for real-time monitoring of RAD by capturing and analyzing airborne sound generated during the fusion phase. Acoustic signals recorded during the process contain noise contributions from multiple machine components, including the transducer and bed movement, which can obscure critical sound features associated with joint formation. To address this, spectral decomposition and deconvolution techniques are employed to isolate fusion-related acoustic signatures from background noise. Initial results show promising correlations between distinct acoustic signatures and different stages of the RAD process, revealing critical insights into joint formation dynamics. This approach enables a more precise assessment of process conditions by successfully filtering out extraneous noise, facilitating real-time defect detection and adaptive process control. The findings establish a foundation for integrating sound-based monitoring as a non-contact, high-resolution quality assessment tool in ultrasonic-based AM. This method enhances process reliability, improves reproducibility, and offers broader applications for monitoring other ultrasonic joining techniques, paving the way for more intelligent, adaptive manufacturing solutions.