RE: [ATM] Achromatic Doublet Design Help

["Jerry" <wa4guu@verizon.net>]

The thickness is largely a mechanical "option". A bit of change in the
thicknesses will not have much effect on the design curves.

	On the positive lens you want the center thickness to be sufficient
to give a good edge thickness to make mounting easy. On the negative lens
you want enough center thickness so that the glass doesn't flex too much in
grinding and polishing. The edge of the negative will naturally be thick
enough if the center is.
	I made a 4" f16 which is negligibly different from f15. Starting
with 1" thick blanks, I made it about as thin as I would go if I were to do
it again. The positive is about .5 inch and the negative about .3 inch thick
at the centers. I didn't have any problem making these thicknesses, but if I
had made the crown much thinner the edge would be getting close to "too
thin". So maybe if you took those numbers and added a bit of a buffer so
that if you had to grind a bit more because of a scratch or something and
ended up thinner than intended, you would not "run out of glass". Around .6
or .7" for the positive and .4 or .5" for the negative wouldn't be bad
numbers to start with.

	You can re-optimize when you get you finished (or should I say
roughly finished?) curves and thicknesses. But you will probably find that
you can't measure them to the precision necessary for re-optimization to be
effective. You will do well to measure your radii to a tenth inch accuracy.
The collective inaccuracies will probably cause the output of the
optimization program to be meaningless. In other words you should end up as
close as you can get to your numbers. If you re-optimize you end up with new
numbers that you will probably not be able to get any closer to than the
last set of numbers. So what is the point. Get as close as you can to your
design numbers. Any improvement will now have to be guided by optical
testing.

You could play with the thicknesses of the elements in OSLO. Vary them a bit
without re-optimizing and see how it affects the OSLO output. You can get a
feel for how much room for error you have. The same could be done with the
radii.

Jerry

 

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