Second attempt at polishing this time with a 14 inch lap on a 3/4inch thick glass base.
A fully polished mirror is a wonderful sight.
Before beginning polishing make certain that the lap is in good contact with the mirror. Move the lap over the mirror, friction should be high and the drag should be uniform. Place the mirror on the lap and move it around slowly, looking through the back of the mirror areas of poor contact should be apparent. The mirror should be polished on an even surface and should not rock during the polishing action, any flexure can result in the introduction of astigmatism. It is helpful to support the mirror on some suitable material, I have used carpet in the past but it absorbs water and has to be washed or discarded after changing abrasive size. With this mirror I used laminate flooring underlay in a double thickness. This material is cheap and does not absorb water. The use of such a support material does not guarantee freedom from astigmatism, therefore I would recommend the mirror be rotated 180 degrees on a regular basis (every 30 minutes). The combination of these two actions should help reduce the possibility of astigmatism.
To polish the mirror I used a mixture of Cerium Oxide and water dispensed from an old washing up liquid bottle. When polishing with a machine (or by hand) it is important to ensure even action across the mirror surface. For the 19 inch mirror I used a 12 inch diameter lap for the initial polish. I doubt that my machine could have pushed a full sized lap back and forth due to the large amount of friction generated by such a lap. There is much to be said in favour of full sized laps, they polish quickly and reduce the chances of astigmatism, but they are difficult to make and even more difficult to work with due to friction.
After 29 hours work with a 12 inch diameter pitch lap I considered the 19 inch mirror fully polished. The surface looked good with one or two sleeks present but these would be of no consequence to the final image. Scratches can look unsightly on an aluminised mirror surface but in reality they have little if any effect on the final image. I am reminded of a true story which happened an at American observatory. Apparently, some lunatic (pun intended) had seen fit to empty a hand gun into the surface of a large telescope mirror, subsequent use of the instrument revealed no apparent degradation in the telescopes performance! I wonder if the offender was a frustrated amateur who finally broke under the pressures of a figuring job?
Testing revealed that my 19 inch mirror had a severe rolled edge which would not respond to my best efforts to shift it. I decided to re-grind from 400 grit using the original 12 inch tiled tool, and re-polish with a lap of 14 inches diameter on a glass base 7/8 inch thick.
The mirror was polished tool on top with a stroke length of 8 inches (just under 1/2 the mirror diameter), with the lap and mirror centres being varied by 2 to 4 inches. The mirror was fully polished after 16 hours (as compared to 29 hours with the 12 inch lap). I used a 10 lb. weight with the 14 inch lap compared to a 5 lb. weight with the 12 inch lap, additional weight and lap size had helped reduce the polishing time. Friction between the 14 inch lap and mirror was at time so high that I had to remove the 10 lb. weight. About halfway through polishing the lap channels were recut using a saw. In between polishing spells the mirror was placed on top of the lap (the two being separated by baking parchment paper to prevent sticking) to maintain good contact.
After one hours polishing the mirror was tested and looked excellent. No turned edge was in evidence, the mirror was smooth with a small amount of correction as would be expected from using a sub-diameter lap. The polish was even across the mirror surface. The mirror was tested at various times throughout polishing. It was found, as expected, that increasing the offset of the lap and mirror centres tended to flatten the mirror, making it more spherical as indicated by straighter Ronchi bands. Reducing the offset between the centres had the opposite effect, adding correction making the ronchi bands more bowed outwards as seen inside the centre of curvature.