Riassunto analitico
Questa tesi presenta un studio approfondito sugli altoparlanti termoacustici (TA) fabbricati con materiali nanotrutturati. La sperimentazione di una nuova tecnica per la deposizione spray di materiali in soluzione ha permesso di fabbricare con successo altoparlanti TA composti da reti casuali di nanofili d’argento e di rame. Per la prima volta i nanofili di rame sono stati utilizzati nella fabbricazione di altoparlanti TA. Un nuovo modello della trasduzione TA, più completo rispetto a quelli presenti in letteratura, è stato sviluppato e validato, fornendo un utile strumento per lo sviluppo della tecnologia. L’altoparlante TA è stato ingegnerizzato sulla base delle informazioni estratte dal modello proposto, culminando nella fabbricazione del primo altoparlante TA con substrato in aerogel di silice. Questo nuovo dispositivo è caratterizzato dalla miglior efficienza tra gli altoparlanti TA con substrato solido presenti in letteratura. Infine, una nuova tecnica di pilotaggio è stata sviluppata per superare due limiti fondamentali degli altoparlanti TA, ovvero la loro intrinseca non linearità e l’elevata temperatura di lavoro. Ciò è realizzato tramite un algoritmo adattativo di predistorsione, implementato con tecniche di elaborazione digitale del segnale. La tecnica di pilotaggio proposta riduce significativamente la temperatura di lavoro dell’altoparlante TA senza compromettere la qualità del suono riprodotto, con prestazioni migliorate rispetto allo stato dell’arte.
|
Abstract
The thermoacoustic speakers are among the firstly-developed electroacoustic transducers, and studied since the beginning of the XXth century. Because of the lack of suitable materials to support an efficient transduction and their intrinsic non-linearity, this technology has long been neglected, leaving the pace to other types of electroacoustic transducers, such as the electrodynamic and the piezoelectric loudspeakers. In recent years the interest in thermoacoustic loudspeakers has been revived following the recent developments of nanotechnologies and nanostructured materials that, characterized by peculiar and unprecedented physical properties, can be used to improve the technology. At the same time, the advanced signal processing techniques allows the response to be linearized. This work shows the modelling and engineering of a thermoacoustic transducer (or thermoacoustic loudspeaker), optimized through the use of nanostructured materials, and the development of the relative control system. The transduction of the thermoacoustic loudspeaker is modelled with a new approach, considering the electro-thermo-acoustic transduction in two atomic and coupled transductions: an electro-thermal transduction and a thermo-acoustic transduction. This approach made it possible to identify the key parameters for the optimization of the device: the thermal capacity per unit area of the active film and the thermal effusivity of the substrate. The effects of the thermal capacity per unit area of the active film have been investigated, revealing its link with the upper limit of the band in which an efficient heat injection in the air is achieved. A low thermal capacity per unit area corresponds to a high cut-off frequency. Prototypes with active films characterized by low thermal capacity per unit area have been fabricated with different types of nanostructured materials and nanometric metal films. The characterization of the prototypes confirmed the model prediction. The effects of thermal effusivity of the substrate were investigated, revealing its link with the efficiency of heat injection into the air. A low thermal effusivity corresponds to a high efficiency. Prototypes with substrates characterized by different thermal effusivities were fabricated using glass and Kapton. The characterization of the prototypes confirmed the model prediction. The developed model has guided the engineering of the thermo-acoustic loudspeaker. An optimized thermoacoustic speaker was fabricated by depositing an active film of 100nm of gold onto a silica aerogel substrate. The fabricated sample is characterized by the highest efficiency among thermoacoustic loudspeakers with solid substrate present in the literature. The linearization of the thermoacoustic loudspeaker response, being static and well known (i.e. quadratic), is obtained by appropriately pre-distorting the desired audio signal with signal processing techniques. Two linearization techniques are presented: the first one, simpler, provides a static predistortion; the second one implements a dynamic pre-distortion to also minimize the power dissipation without affecting the signal reproduction.
|
