ATM LIttle Glass Buttons

[Aplanatic@aol.com]

Here's a problem I've been pondering all morning.  Perhaps you can help.

I have two pregenerated optical glass blanks that will be used in an 18" hyperbolic astrograph.  They are F2 and BK7, 4.4" and 5" in diameter respectively.  Both sides of each blank will be ground and polished to a sphere.

I've poured mating dental stone substates for each of the four surfaces and epoxied small washers onto each substrate.  Initial grinding experiments with the washers are not encouraging.  The washers work fine as a grinding tool but are so slow to wear that it's nearly impossible to change the radius of curvature of the lens surfaces [1].  So, it's time to pour four new substrates and line these new substrates with something that has a wear characteristic closer to glass, i.e., glass or ceramic tile.

Obviously, I could just start cutting up some 1/8" thick plate glass into 1/2" squares, or could buy some 1" ceramic tiles and cut them into 1/2" squares [2].  That seems like a lot of trouble.  

So, the question is this:  Where can I find relatively thin (1/8" or so) glass or ceramic disks that are on the order of a 1/2" in diameter?  How about 1/2" squares?  Any other interesting suggestions?  

Thanks in advance,
Dave Rowe

[1] Glass Fab, Inc. did a beautiful job on the blanks, pregenerating the surfaces to my specs within about 2%.  However, for this application I must get within 0.25% accuracy for the RoC of two of the critical surfaces.  Clearly, I need a tool that will allow me to easily fine tune the RoC of all four surfaces.  Steel washers are not appropriate for this application, but should work fine for less stringent grinding requirements.

[2]  The radius of curvature of two of the surfaces is approximately 8".  The sagitta of a 1" square is rather large for this RoC and would require a lot of grinding to conform to the pregenerated blank.  Clearly, a 1/2" square "tile" is the right size for this application.