RESONANT CURVED PIEZOELECTRIC CANTILEVER FLUID DIODE WINGS FOR MASS-PRODUCIBLE FLYING MICROROBOTS Theses uri icon

  •  
  • Overview
  •  

abstract

  • This work explores a new method of force generation for flying robots on the sub-cm wingspan scale: resonant curved piezoelectric cantilevers created using completely parallel MEMS fabrication. It theorizes that because a resonating curved beam has a different drag coefficient on the upstroke than the downstroke, it should act as a fluid diode: a partial one-way gate for fluids, and thereby generate an asymmetric force over a symmetric one-degree-of-freedom flapping cycle. It develops a simplified model for the large-amplitude resonant mode of thin circular arcs by analytically extending the resonant mode shape of straight cantilevers, shows that this shape is a better fit to experimental data than previous models, and shows that it accurately predicts the resonant frequency. It uses this resonant mode to compute the force on flapping curved arcs under a wide range of amplitudes, Reynolds numbers, and arc angles using computational fluid dynamics (CFD) simulations, and extends the concept of a drag coefficient from steady-flow fluid mechanics to steady-state oscillatory fluid mechanics both for net force generation and power dissipation. It develops a framework to analyze the CFD results in the broader context of a complete robot, and uses this framework to determine priorities for material selection, robot size, and flapping shape, depending on desired robot application. It tests these theoretical predictions by creating prototype 7.6 mm wings out of 7.5 micrometer thick x-cut quartz and SU-8, after developing and implementing a method to smoothly thin x-cut quartz leaving the surface free of dielectric-compromising pits using reactive ion etching (RIE). Finally, it constructs a test chamber to measure the force, amplitude, and electrical parameters of the flapping wings under a variety of air pressures and demonstrates that the results are consistent with the theoretical predictions, indicating that this approach can in fact lead to successful flying microrobots.

publication date

  • April 2014