Re: [ATM] FW: Figuring flats

[Mike Lockwood <melockwo@uiuc.edu>]

Philip,

Philip Lardner wrote:
> Thanks for the quick response. Yes, all three disks are plate/float
> glass, 1" thick. Their backs were not ground flat prior to fine
> grinding the fronts because I didn't fancy having to polish them
> out too! Perhaps I should have. However, judging by the ware
> pattern as I fine ground the fronts, they were only very slightly
> out of flat, producing a slight saddle-pattern between the ground
> and un-ground areas. This disappeared very quickly. I am using a 
> 1/2" thick foam-rubber camping mat to support the work on top of my
> work stand (because it is water-proof and washable) and a piece of
> carpet, covered in black cloth, to support the plates in the
> tester.

That saddle pattern is all too familiar.  In my thin-mirror 
experience, it will still tend to print through from the back of the 
disk, even though you have the disks on a flexible support.

If the backs are now a figure of revolution (de-saddled) then this 
will help with the figuring process.  It probably changed the shape of 
the optical face if you ground them.  I'm guessing they got more concave.

I'd grind the backs to 12 micron if you can, and then put on a quick 
polish so you can see fringes through them.  As for that polish....

Mitch wrote:
> With this talk of polishing the back, why not use a "brute force" 
> method of a car waxer/polisher?

It works very well, so long as the mirror is held in cleats and will 
not be flung across the room by the polisher!  It's a great way to 
make the mirror transparent.  I have also oiled the back of a flat to 
see through it, and that works OK, too, but it is messy.

Just for completeness, removing the stress from the blank requires 
removal of all of the subsurface damage (cracks) and this requires 
just as thorough of a grind as the optical surface.  I would not go to 
this trouble for plate glass flats - after you have gotten them fairly 
close, then I recommend you make your final "reference" flat out of a 
thicker piece Pyrex (10"x2" is nice) that you have carefully checked 
the anneal of.

> I have a 14" f/15 sphere (the primary mirror for the folded
> Stevick-Paul) that can be pressed into service for a Ritchie-Common
> test. The only problem with that is having to build a test tunnel
> 10.5m long to control the air currents. It is devilishly sensitive
> to air movement... and I live in a big old draughty house... with
> central heating at this time of year! I will *try* and knock up a
> Ritchie-common test and capture some images this weekend. I'll also
> have a play with FringeXP.

Once you have the backs flat, do some more polishing of the optical 
faces while trying to keep their figures under control.  It may take a 
while to remove the astigmatism, but it will go away eventually.  Then 
your test results should be more consistent.

You need to learn what astigmatism looks like, in terms of a fringe 
pattern.  In general it is the non-parallelness of the fringes that 
gives it away.  Become sensitive to that.

	Mike L.

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