Maximum Effectiveness of Electrostatic Energy Harvesters When Coupled to Interface Circuits

Many motion driven energy harvesting devices using the electrostatic force have been demonstrated, with many of them using a moving plate, variable capacitor structure. The output power of all reported electrostatic energy harvester systems that interface to an energy storage element is at least one and sometimes more than two orders of magnitude below the theoretical maximum limit for the given device dimensions and driving motion. This paper shows that the theoretical limits on the harvester performance when only the coupling effectiveness is considered are misleading and that when the transducer and its power processor interface are combined, further important limitations apply which significantly reduce the theoretical maximum power. The combined analysis rests on a parameterization of the inherent and parasitic properties of key components, notably the power semiconductor devices. Although scaling laws in general favour electrostatic force solutions over electromagnetic force solutions at micro scale, the specific compromises encountered with the interface circuits for electrostatic generators means that their range of applications needs to be re-examined. Under previous analysis, the choice of constant charge or constant voltage operation has not been fully resolved. The new analysis shows that when the power electronic interface is considered together with the transducer, the performance of the constant charge harvester system is generally poor, although an acceptable region of operation limited to intermediate sizes and accelerations exists. The constant voltage device can operate with acceptable effectiveness over a much wider envelope, and is thus the preferred implementation. The optimization of the transducer and interface circuits underlying these conclusions was performed in MATLAB and verified with time-domain PSpice simulations.