The closed-loop dynamics of a chaotic electrostatic microbeam actuator are presented. The actuator was found to be an asymmetric two-well potential system with two distinct chaotic attractors: one of which occurs predominantly in the lower well and a second that visits a lower-well orbit and a two-well orbit. Bifurcation diagrams obtained by sweeping the ac voltage amplitudes and frequency are presented. Period doubling, reverse period doubling, and the one-well chaos through period doubling are observed in amplitude sweep. In frequency sweep, period doubling, one-well, and two-well chaos, superharmonic resonances and on and off chaotic oscillations are found.
We experimentally investigate the primary superharmonic of order two and subharmonic of order one-half resonances of an electrostatic MEMS actuator under direct excitation. We identify the parameters of a one degree of freedom (1-DOF) generalized Duffing oscillator model representing it. The experiments were conducted in soft vacuum to reduce squeeze-film damping, and the actuator response was measured optically using a laser vibrometer. The predictions of the identified model were found to be in close agreement with the experimental results. We also identified the noise spectral density of process (actuation voltage) and measurement noise.
1 March 2013 in Transactions of the Canadian Society for Mechanical Engineering
Natural frequencies of a clamped axially loaded beam carrying an eccentric end rigid body are computed. A linearly varying non-follower axial force representing the beam’s own weight is taken into consideration. An analytical form of the frequency equation of the structure is obtained and solved numerically. The parameters associated with the end rigid body, the mass, the rotary inertia and the eccentricity, are shown to considerably affect the natural frequencies of the structure. The modified orthogonality conditions of the system are presented. A closed-form of the frequency equation for a beam under constant axial load carrying an eccentric end rigid body is obtained and results are compared with the distributed load case. Based on the exact and approximate computations of the critical buckling load a recommendation is made on the method of including the effect of distributed load for practical purposes.