The use of voice coils to directly drive scanning at the desired velocities has been ruled out on practical grounds. Rotary motion may be a more appropriate method to obtain very high mirror velocities. There is, however, a symmetry mismatch between the nominally planar wavefronts of a Michelson interferometer and surfaces of rotation. A rotating optical element would probably have to spin at 500 revolutions per second (rps) to achieve the desired 1 millisecond scan time. Many issues would arise in such a design, including turbulence around the moving optics, vibration and mechanical stress.