From: Aplanatic{at}aol.com
To: tomjan{at}ozemail.com.au ("Thomas Janstrom"), atm{at}shore.net
Reply-To: Aplanatic{at}aol.com


Hi Thomas,

As you may have found, the convex mirror and its test-plate mate tend to
become mutually concave during polishing.  Interpretation and analysis of
the fringe pattern becomes difficult when too many fringes are present.

To keep the two pieces of glass close to the same radius of curvature,
polish the convex mirror with tool on top and/or polish the test plate with
the tool on the bottom.

I've tried using three pieces of cellophane between the disks and found it
to be very annoying and hard to remove the air wedge with this method.  I
now test with the two optics in contact and don't use spacers between the
two.  This can be a bit dangerous if the glass is not very clean because of
the risk of picking up scratches, but it makes it much easier to remove the
air wedge.

Other notes:

Let the two disks thermalize before making any serious measurements.

There will surely be some residual curvature mismatch between the disks. 
You will have to take measurements of the fringe locations and remove the
second-order radial dependence in order to determine the fourth- and
higher-order deviations from your desired surface profile.  It's only these
higher-order deviations that make any difference to the optics.

If you do use spacers between the disks, make sure you account for the
built-in path length differences when the light source is not at
"infinity".

Figuring the convex secondary can be frustrating.  As an earlier post
noted, it's often useful to adjust the lap by symmetrically removing pitch
from some regions.  Be slow and careful with this technique, it can be
quite agressive.

Best of luck.

-Dave Rowe-

> Ok I now have my DK secondary ready to test, at a first glance (no wedges
> used) I get about 26 fringes neatly centered. With wedges I get arcs, and I
> mean lots of them, far too many to count. I have tried several wedging
> options, some with as small a difference as 2.5microns, the largest was
> 0.1mm, in both cases there were more fringes than you would want to have to
> count.
>
> What should I do to bring the fringe numbers down (this question is only RE:
> using an air wedge) Tex suggests a shallow wedge, but I tried that....
>
> Secondly what is the recommended polishing technique to
> steepen a convex
> curve?
> Clear skies, Thomas
> http://www.tjanstrom.com
> "Don't make me set the laser printer to stun"

--- BBBS/NT v4.00 MP