Given the fact that you spent as much time as you did polishing, I would bet that your methology in GRINDING is to fault. Don't rush moving from one grit to the next. Slavishly look for remaining large pits and keep at it until you remove them.
> I did tie him down, and he said "good" is a maximum of 1/4 Wave
> wavefront error, so a maximum of 1/8 Wave surface error since they use a
> two-pass system. (Whatever that is)
> -Alan Kilian
>-- End of excerpt from Alan Kilian
Sounds like he is testing your mirror using autocollimation against a flat. In such a test setup, a light source is placed at the focus (not the center of curvature) and light spreads out to hit the mirror. It then bounces off of a flat and returns to the focus point where it is cut by either a knife edge or a Ronchi screen. Under these test circumstances, the parabola should look exactly like a sphere. Because the light bounces off the mirror twice, it looks twice as bad as it really is.
As for masking off your mirror, just take a Sharpie and put your mirror on a turntable. Black the outer 1/8-1/4". (Don't scrimp on your pen. Make sure it is a permanent ink pen. You don't want dew making it run like mascara all over the face of your mirror).
--
Mark T. VandeWettering Telescope Information (and more)
Email: <markv@pixar.com> http://webspace.com/markv/
<markv@webspace.com> Clear Skies!