Elastodynamic analysis of langevin transducers and sonotrodes for high-power ultrasonic processes

The thesis focuses on the methodological approach proposed for the development and validation of finite element models and multiphysics simulations of Langevin-type ultrasonic transducers. These transducers, employed in ultrasonic processing systems, require increasingly high performance and must comply with stricter requirements in terms of repeatability and durability. They are subjected to complex electromechanical interactions and long-term aging phenomena that affect their reliability and operational efficiency. This work aims to provide a CAE tool featuring improved accuracy in the numerical prediction of Langevin transducers’ performance under different operational conditions, i.e., variable process parameters and healthy/faulty states of their components. The results of numerical simulations can be exploited to optimize the design of components as well as to support the analysis of experimental measurements acquired for condition monitoring and diagnostics of in-service devices. In particular, a novel modeling strategy to address the effects of bolt preload on the electro‑elastodynamic behavior of transducers is proposed. Unlike traditional approaches that apply preload as a loading boundary condition, the proposed method adjusts contact stiffness between stacked components, enabling a more accurate representation of mechanical coupling and its influence on system dynamics. A significant part of the research involves an experimental campaign aimed at characterizing the time‑dependent evolution of electromechanical properties in Langevin transducers and individual piezoelectric rings. Aging effects, such as variations in impedance curves, electromechanical coupling coefficient, current gain, dielectric and piezoelectric constants, were analyzed under different excitation conditions. A heat‑treatment procedure was developed and validated to accelerate stabilization, reducing the aging period from several months to a few days. By integrating validated numerical models with experimental insights, this work provides a robust framework for design optimization, performance prediction, and health monitoring of ultrasonic transducers. The findings enhance understanding of aging mechanisms and offer practical solutions for improving dynamic response prediction and diagnostic evaluation.