Broadband acoustic energy harvesting via synthesized electrical loading

Abstract

The need for self-powered wireless sensor nodes is ever increasing. One promising technology for self-powered sensor nodes is acoustic energy harvesting (AEH): deriving energy from ambient sound. Current AEH designs are typically based on resonant structures, yielding narrowband energy harvesting and therefore low efficiencies from broadband noise sources. They also generally exhibit MEMS-scale sizes, with consequently low power outputs. This work addresses the size and bandwidth of AEH devices. A large-scale acoustic energy harvester is presented, based on piezoelectric Polyvinylidene Fluoride film, 100cm2 in size. This harvester design was selected after analysis and comparison of magnetic, electrostatic, piezoelectric, and triboelectric transduction. An energy-based dynamics analysis of such design a yields a third-order nonlinear differential equation, modeling the electromechanical dynamics of the system in open-circuit conditions. The model can be represented by a linearized equivalent circuit and subsequently a Thevenin equivalent model. Optimal broadband energy harvesting is achieved in theory with a conjugate matched load at all frequencies. This load is realized using operational amplifier circuitry, with special attention paid to stability challenges. The AEH design was fabricated and tested with acoustic input over a range of 70Hz-7KHz. The model was validated experimentally via open-circuit voltage measurements and delivered power measurements with resistive loading. The AEH design was loaded with the designed conjugate matched load, with corresponding experimental voltage and delivered power measurements to demonstrate power output and bandwidth improvement. While stability challenges and sensitivity to load capacitance precluded a perfect impedance match, broadband performance was achieved exceeding that possible with purely resistive loads, or with resonant structures demonstrated in literature. The implemented design harvests 1.6uJ per takeoff event of a 747 aircraft, or 0.25% or available power, requiring 58 volts to generate the forces necessary for impedance match. A perfect impedance match of this would require 1387 volts, harvesting 491uJ per takeoff event. Losses arise primarily at low frequencies, where a poor impedance match exists and significant energy exists. Given resolutions to stability, sensitivity and voltage challenges, the technology has the potential to be scaled up further and used in additional applications such as large-scale sound absorption.