[ATM] Spherometers and sphericity

[Mike Lockwood <melockwo@uiuc.edu>]

Pete,

ArtfulBodger wrote:
> Michael, David, um ... a spherometer isn't good at testing 
> sphericity?  How come?  Do you mean the fact that it can't check near 
> the edge of the mirror?  I'm guessing that that's what you have in 
> mind, but maybe not.  Could you (or someone) elaborate, please?

If you have a spherometer with a sensitive indicator (reading to 
0.0001" or better) then you can measure zones left in fine grinding by 
sliding it around.  The indicator will change its reading if the 
mirror is not spherical.  With ball bearing feet and a rounded 
indicator tip, I have never gotten scratches from doing this.

You CAN measure near the edge of the mirror - move the spherometer so 
that one of the feet gets near the edge and watch the indicator.  If 
it moves, something may be going on there.  (The chances of having a 
zone under the other feet are smaller than at the edge.)  While the 
sensitivity is reduced, you can still see the change as the outer foot 
rises or falls.

Sliding a spherometer that has a leadscrew-type micrometer is unwise, 
since it either contacts or doesn't contact the surface.  Not a good 
idea to slide something around on your mirror with a metal probe 
rigidly protruding until it "catches" on a high spot.....

> About the Sharpie test, I can still do it with the finer grits, can't 
> I?  25u, 12u, and like that?  I guess the mirrors should be spherical 
> before then, and they were, but I'd like to check 'em again (and 
> again).  Make sure they're good & round before I start polishing. Can 
> I do that?

It works fine down to 25u.  After that it can interfere with grinding, 
depending on the ink thickness.  Be gentle with 12u or 5u if you have 
ink on the mirror.

The sharpie test is not an absolute test for sphericity because it 
merely shows where the grinding/wear is happenning.  Abrasive layers 
and grinding action can be, and usually are, non-uniform.  Randomizing 
strokes (or moving them around on a machine) are ways to make it 
average out to uniform.

The pencil test is another option.  Put lines on the mirror with a 
pencil, then gently rub it against the tool WITHOUT any abrasive.  The 
two are in contact where the pencil lines go away or are disturbed. 
Again, this does not guarantee sphericity, but it's a fairly good 
indicator of it.

	Mike Lockwood


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