Re: [ATM] Spherometers and sphericity

[Dominic-Luc Webb <dlwebb@canit.se>]

On Wed, 28 Nov 2007, Francis J. O'Reilly wrote:

> Guys (and ladies if there are any on the list):
>
>
> I must be seriously out of touch. I never do any testing of the surface
> whatsoever until I am fully polished out.

For typical mirrors in the F/6 and higher range and following the
basic rules, this usually works fine. Checking for a sphere gets
more important as you get into really fast mirrors because you
can easily get quite far from a sphere even during coarse
grinding, and you obviously would not want to have to go back
to coarse grinding a really deep curve. With a spherometer or
other methods, you can determine that you have a more or less
spherical surface pretty much all the way to polishing and if
you stray away, you will catch it early enough that you won'y
end up losing a lot of time.

After having completed a few slower mirrors where I actually did
have a sphere right up to polishing without needing to check, I
reckoned I was able to make my F/1.0 and F/1.55 mirrors with
checking. Actually, the F/1.0 came out almost perfect, but I
did need to go back to 17 micron SiC. The slumped F/1.55 was
a bigger problem, and I suspect this was a combination of flexing
of such thin plate glass (about 19 mm thick, 310 mm diameter)
plus the deep curve. Building a cell first and doing the grinding
with mirror in this cell, plus regular checking resolved most of
the problems, and the rest I have solutions for.

Dominic

_______________________________________________
ATM mailing list http://www.atmlist.net/