A proposed lightweight, micromachined, multiaxis-steerable mirror would have mesoscopic dimensions. Its steering function would involve mesoscopic positional excursions of its support points and would be characterized by rapid slewing and moderate angular precision.
The mirror, its supporting structures, and its steering mechanisms would be fabricated in a four-layer polycrystalline-silicon surface-micromachining process. The mirror would be triangular and would be supported at its corners. The steering mechanisms and supporting structures connected to two of the corners would include electrostatic comb drives, reduction gear trains, gear-driven translation racks, hinged beams, and flexures (see figure).
In one example design, the electrostatic comb drives would operate in four-phase cycles. For each such cycle, a 19-tooth pinion gear would undergo one revolution. The thrust of each comb drive would be ≈1 µN and could be applied rapidly enough to sustain rotational speeds up about 4,000 revolutions per second. Two tandem gear trains would reduce the translation to 1.66 µm per pinion rotation or, equivalently, 0.415 µm per phase step of an electrostatic comb drive, and would increase the torque proportionally.
The triangular mirror would have sides 1.3 mm long. Translation of one corner of the mirror in a 0.415-µm increment as described above would result in a mirror rotation of ≈0.3 milliradian. The flexures would serve as smooth (free of stiction) means of bending and rotational coupling between the corners of the mirror and beams that hold the mirror on a substrate. The flexures could accommodate rotations and bends of at least half a radian. The maximum speed of translation, of a hinged beam would be 6 mm/s, corresponding to a rate of rotation of about a radian per second.
In principle, the mirror-aiming direction would be a known function of the number of phase steps from starting positions of the electrostatic comb drives. Inasmuch as the out-of-plane bending and torsion of the flexures would generate opposing forces on beams and all the way back to the pinion gears and comb drives, there would be no gear backlash and thus no need to account for backlash in calculating or controlling the mirror-aiming direction.
This work was done by Frank T. Hartley of Caltech for NASA's Jet Propulsion Laboratory.
This Brief includes a Technical Support Package (TSP).
Unfortunately the TSP Mesoscopic Steerable Mirror (reference NPO-20971-) appears to be missing from our system.
Don't have an account? Sign up here.