On Thursday, October 25, 2012 9:14:48 AM UTC-7, Phil Hobbs wrote: [about rotating-coil magnetometers]
Two other approaches: One: wind the coil with one bit rotating in the tip, coil-axis transverse to rotation, and another coil (in series) that's far from the tip and coaxial to the rotation, with a surrounding or axial-adjacent pickup coil. Main problem: shielding the output coupling coil.
Two: use a battery on the rotating part, and run a CD4046 with some DC bias as a current-to-frequency converter, with output blinking a LED, and pickup with photocell and another CD4046 as frequency-to-voltage receiver.
A couple of lithium coin cells is cheaper than good sliprings.