[ATM] RE: ATM Schiefspieglers in Oslo LT- distortion plot problemsolved

[Mauritz Andersson <mauritz@physi.uni-heidelberg.de>]

It seems that OSLO has problem with the reference points.
Go to the "Ray Analysis" menu in the graph window meny and
then press the "Distortion Plot" button. There the distortion
grid plot to the left shows that the distortion is insignificant,
whereas the lower right plot shows the absolute distortion
in microns and there is clearly something stange since it quickly
goes to a constant value close to 38000 mu (1.5 inch)!

To circumvent the problem one can use a different way of tilting the
mirrors, namely a tilted coordinate system via "tilt and bend". 
In the coordinate data input one can select to use instead
a "tilt and bend" and the the subsequent mirrors are relative to
this new coordinate system. Thus each mirror need only to be 
tilted and should not be displaced.
See lens file
http://www.physi.uni-heidelberg.de/~mauritz/Stevick-Paul8_tiltandbend.len

If you now do the distortion plot it is more realistic.
The other advantage with tilt and bend is that one always follows
the optical axis which is the relevant axis for measuring the image
plane tilt (approx 9 deg for this design).

Anyway, the lens is excellent, Don.

BTW, did anyone manage to do a spot diagram plot on-axis
and off-axis in all four directions like in the TCT program?
Since tilted telescopes usually do not have equal off-axis
abberations in each axis this would be nice to be able to plot.
I have not found an easy way but maybe someone wrote a little ccl
code that does the trick or know of some way to do it?

Best regards,
Mauritz Andersson


On Fri, 2004-06-11 at 05:57, Donald Good wrote:
> I think that items that are calculated relative to some unknown reference
> such as indicated for the distortion reference height might be suspect.
> Items that are calculated on intrinsic characteristics (ray tracing based on
> optics geometry) probably can be relied on to the extent that program errors
> are minimized.  Again, as the documentation stated, "it is important for the
> user to know what the program is computing."  I think that if the values of
> items that you understand agree and indicate superior performance, then
> those that you don't understand can be ignored, especially when the
> documentation gives warnings about the applicability or relevancy of those
> items as it did twice about the distortion plot.  If I had reasonable doubts
> about the unusual distortion plot, I would investigate further.  As it is, I
> don't think it is worth the time.
> Clear skies,
> Don

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