|
Ken, aha - there is a way - and it should be quite good.
Assuming you have (i) a large concave telescope primary mirror and (ii) the gear to do a Foucault or Ronchi test, set this up using the secondary mirror flat as you would in a double-pass autocollimation (DPAC) test:
Light from light source to concave mirror, reflected to the flat and back to the concave mirror then reflected back to knife edge or ronchi screen. Knife edge/ronchi should be close to the light source, and both at the centre of curvature of the concave mirror.
Although technically you need a concave parabolic mirror, the aperture of a secondary mirror is much smaller so only a small portion of the concave is used, ie it will effectively be working at a very long focal ratio. Defects in the primary such as spherical aberration won’t matter much with one exception - turned down edge - and in this respect as long as the light path avoids the edges and utilises the central section of the concave mirror, it should be happy days.
You could also test the primary beforehand to identify which areas are the best ie smooth without significant localised surface errors.
Defects seen in the DPAC test will be due to the secondary, and as only 1 bounce is used off that, wavefront error will be as measured (not half as when DPAC is used to measure a scope or primary mirror).
|