Re: [ATM] Malacara, Flats etc... topic was "They come through for me, again"

[Richard <cnc@cncservo.co.uk>]

Hi Chris,

Wednesday, February 21, 2007, 4:55:33 PM, you wrote:

CDP> To avoid the need for a flat, I measure the difference between my tool and
CDP> mirror using my spherometer and divide by 2.

CDP> I've been figuring that any inaccuracy due to imperfect contact such as an
CDP> gap in the center or an gap at the edge would be small enough to ignore.

CDP> I'm interested to know, how does that sound to everyone?

Personally, my first reaction was to look at a calendar to see if was
April 1st. (Added later - but then I did think you were trying to make
a flat with a spherometer!, sorry).

If you are Superman in a cleanroom, you are going to get at best +/-
0.001mm on a transfer measurement. That's around 2 wavelengths at the
surface, 4 at the wavefront. A flat usually being a reference element,
that's at least 40 times out of spec.

I do actually have a super accurate linear transducer that will
measure to 0.0001mm *resolution* with an accuracy of around 0.0003mm
but trying to actually achieve anywhere near that under real measuring
conditions is almost impossible, even with averaging.

With the usual interferometric techniques you can just look at
discrepancies much less than that with your eye.

-- 

Best regards,
  Richard in the UK

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