Image of the day

From the
ATWB Customer Gallery

Eskimo Nebula

My Account

New to Astromart?

Register an account...

Need Help?

fine grinding

Started by Phil Barker, 04/05/2013 10:06AM
Posted 04/05/2013 10:06AM | Edited 04/05/2013 10:08AM Opening Post
Over the years I've polished a few mirrors and the subject of fine grinding and ensuring you have close to a spherical figure before polishing has perplexed me a lot.

I have a bit of a theory about to get it right and basically when I get down to 25 micron down I fine grind for around an hour then do a test with a permanent marker.

I draw a grid pattern and then apply some 25 micron paste and move the mirror in different directions to work out of the figure is spherical.

If its not in contact in the middle the solution is to use the tool on top and test after some work and then work out if its closer to a sphere.

I then look for pits using the sun. I need glasses nowadays to do this correctly but this very effectively shows up any pits from a previous grade of abrasive. These must be removed before I go to the next stage.

Once I have good contact and I'm free of previous pits I go to 9 micron. This is a big step and I usually do 2 hours with this repeating the process above. I also use the sun or a bright filament of a lightbuld on an angle to see if the surface if even from centre to the edge.

I've noticed for example if the centre if not in good contact as shown by the sharpie I can easily see that the image of the filament fades out in the centre compared to the rest of the surface.

Good contact,checking the reflection on an angle and looking for pits always means I know I can confidently go to the next grade. I usually finish with 3 micron and have successfully gone from 9 to 3 using this technique.

At 3 micron If I have good contact no previous pits and the view on an angle is good I can be confident polishing will go well.

In the past I've seen things like greyness in the centre or edge and its because I havent had good contact. Grinding removes glass a lot faster than polishing and sometimes its better to bite the bullet and regrind. I find that I usually have to go back to 15 micron or 25 to ensure I have ground out any greyness and then follow the process of ensuring I have a sphere described above.

I'd be interested in any thoughts about this and if others agree or not. Presently I'm doing a crown element for a refractor and on 5 micron on R2.

Posted 04/05/2013 08:52PM #1
Hi Phil,
The method of figure control you mentioned works and in combination with a drop gauge for measuring CT you can assure removal of subsurface damage from prior abrasives.

A lot of work was done by the high power laser folks and industry in the early fusion optics days. From this work most shops developed minimal material removal protocols that are typically based on some assumptions of fracture depth for the prior abrasive or machining operation. To confirm these subsurface damage depths an acid etch using HF or Ammonium Biflouride followed by an examination with a surface profiler was used and-or the optic was sectioned and examined with an SEM. There is another method using TIR form a right angle prism hypotenuse that works as well… Very long story short there are minimums you can use to assure complete removal of subsurface damage from successive loose abrasive grinds. There are also equations you can apply for materials of varying fracture toughness.

Old style generators leave damaged surfaces akin to rough loose abrasive grinding we use for making our telescope optics. And in the early days those generators were the only thing available. Generating was followed with a 20 micron loose abrasive grind and removed ~0.005” of material. This was followed with a 9 micron loose abrasive grind where 0.003” was removed, a 5 micron grind was used and 0.002” were removed and the part was then polished. There are other nuances most high end shops consider trade secrets, but that’s the just of it. Softer materials are often scratched with 5 micron so we polished directly from 9 micron. It is very easy to pick up scratches from 5 micron and finer abrasives so most of the shops I am familiar with stopped at 5 micron. An operator running many spindles cannot give the attention needed for really fine abrasive grinding.

This protocol is likely over kill for your primary mirrors but is offered as a starting point. Good control of figure, using a method like you described, in concert with measured material removal will assure damage removal.

George Blahun said: “I’d just keep polishing and keep track of the quality with a laser pen on the front surface, there should be no spot on it when fully polished.” An excellent suggestion and exactly akin to what the better shops do when evaluating polished surfaces…

As a side note: Today’s precision generating machines don’t require any cleanup grinding if run correctly. Parts coming off a well run precision generator looked polished, and small lens surfaces can be final polished and figured using pads in a matter of minutes with figures better than ¼ wave.

Best regards,
Dave