School of Mechanical Engineering: Oren Gad Levi and Ben Torteman

~~

SCHOOL OF MECHANICAL ENGINEERING SEMINAR
Wednesday, December 11, 2019 at 14:00
Wolfson Building of Mechanical Engineering, Room 206

Sub-g Bistable Frequency Sensor with a Tunable Threshold

Oren Gad Levi
MSc student of Prof. Slava Krylov
Vibrations monitoring lies at the foundations of structural and industrial machinery health monitoring, autonomous wake-up sensors for Internet of Things (IoT), or even peacemakers. These applications impose demanding requirements of low power consumption, high accuracy, and real-time measurement capabilities within a wide region of frequencies often down to several Hz. While approaches based on microelectromechanical (MEMS) accelerometers or optical fiber Bragg grating (FBG) sensors and involving electronic post-processing of the time-history data are well established, required computing power and energy consumption are often incompatible with the emerging applications demands.
In this work, we demonstrate, using a model and in an experiment, the functionality of an electrostatically actuated bistable electromechanical sensor allowing real-time, wide-band measurement of a macro scale host structure frequency. The operational principle of the inertial switch with an adjustable threshold is based on the level crossing approach. Electrostatically actuated devices are prone to the so-called pull-in instability when the structure collapses toward the electrode at the actuating voltage exceeding certain critical value. In the presence of an inertial signal, the pull-in voltage is affected by the acceleration. In our experiments, a 4 mm × 4 mm device fabricated from a 50 m thick metallic foil was attached to a vibrating host structure. The signal frequencies in the range between ≈ 0 Hz and up to ≈ 900 Hz were measured by monitoring the time intervals between the pull-in events, registered as an ohmic contact between the electrodes. Operation in the vicinity of the critical pull-in point enhanced sensitivity and allowed measurements of sub-g accelerations. Since the device combines both sensing and signal-processing functions, the post-processing of the output, which is wirelessly transmitted to a cell phone, is reduced to minimum. The device was used for the frequency measurement of a realistic macro-scale structure and the frequency accuracy of   0.16 % was demonstrated.