Towards a high bias voltage MEMS filter using electrostatic levitation. (March 2021)

• Record Type: Journal Article
• Authors: Pallay, Mark; Miles, Ronald N.; Towfighian, Shahrzad
• Is Part Of: Mechanical systems and signal processing. Volume 150(2021)
• Publication Date: 2021-03
• Subjects: MEMS -- Filter -- Electrostatic levitation -- High voltage; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2020.107250
• Languages: English
• ISSNs: 0888-3270

Abstract

Highlights: A MEMS filter can function using electrostatic levitation for sensing and actuating. The filter can withstand very high voltages without becoming unstable. A high voltage boosts signal-to-noise ratio without hindering mechanical response. The large actuating voltage improves the mechanical performance of the filter. Abstract: Traditional MEMS filters use a comb drive structure that suffers from the pull-in instability, which places a significant limitation on the achievable signal-to-noise ratio of the sensor. Because the output signal from a capacitive sensor is linearly related to the applied voltage, it is desirable to use a capacitive sensor that can withstand large voltages upwards of 100 V. However, the pull-in instability causes high voltages to destroy the device and a trade-off between performance and reliability must be made. Electrostatic levitation, which works by pulling electrodes apart instead of together, eliminates the pull-in instability and allows for very high voltages to be applied without damaging or destroying the sensor/actuator. This study theoretically and experimentally demonstrates that a filter based on electrostatic levitation eliminates the voltage limitation of the capacitive sensor, which has historically hampered the performance of the filter. A model of the filter is derived and validated with experimental data. Voltages up to 100 V are applied without damaging the filter.