Re: [ATM] My first sphere

[mdholm@telerama.com]

Hi Darran,

If you are sure the pits are all polished out, then it is time to start figuring.

First, go to very much shorter polishing sessions. 5 - 10 minutes at a time,
moving to even less as you get close to the final goal.  Also, press well
between sessions.  Poor contact can drive you nuts with unpredictable results. 
Keep using the netting.  In my experience (and that of others), it leads to
smoother polishing, better control and better results.  After you pull the net
off, press a while longer to mash down any pitch raised up by the removal.

Make your stroke slow and even.  You are no longer trying to wear away lots of
glass to get through pits.  Now even, slow, steady, controllable removal is the
goal.  If the lap gets grabby, causing uneven motion, stop and repress with the
netting.  Don't go on polishig with a grabby lap, you will get uneven and
unpredictable results.  (I learned that the hard way!)

Looks to me like you have a hill in the middle half of the mirror and a turned
up edge.

Try a few minutes of parabolizing stroke.  A reasonably wide W stroke, MOT.  Not
more than a few minutes, and then look again.

If I were doing it, I would start making Foucault zonal tests.  I like the
quantitative results.  Some would say not to bother until you get a sphere, but
I like to see that the Foucault numbers agree with what my eyes are telling me.
 I have also found that the Foucault numbers will show changes that I may be
uncertain about by just looking.

Also, for a first timer, a bit of introductory practice with the Foucault test
will be valuable.  Better to learn now while you are still a little way from
making those final judgement calls on a nearly done mirror.  If you really are
moving toward a sphere, all the Foucault numbers should contract towards zero.

A slightly discomfiting thing is that, with the high center and turned up edge,
some of your Foucault numbers are going to be negative.  That is OK, it just
says the Foucault test is also seeing the high center and edge.

Some atm's like to judge mirror shape just from the raw Foucault numbers.  I
prefer to let Sixtests do it.  Jim Burrows is solid on the math, and has
implemented about the most sophisticated math that can be brought to bear on
Foucault results.  It is much less likely to make an interpretation error than I.

I don't know if you are planning to use the Foucault test, or the matching
Ronchi test.  I am not able to give good advice for matching Ronchi, since I
have never used it.

For the Foucault zonal test, you will want a 4 zone mask.  You can calculate the
zones using Nils Olof Carlin's web calculator
http://www.atmsite.org/contrib/Carlin/couder/  or use my program CouderMask.exe
(for Windows 95 and higher) to calculate and print the mask pattern 
http://www.atmsite.org/contrib/Holm/coudermask/couderma.html

My program uses exactly the same math as Nils Olof's web site.

For analyzing Foucault zonal test results, I strongly recommend Jim Burrows'
program Sixtests 
http://home.earthlink.net/~burrjaw/atm/atm_math.lwp/atm_math.htm You want the
Windows version.

Mark Holm
mdholm@telerama.com
_______________________________________________
ATM mailing list http://www.atmlist.net/