On Wed, Dec 7, 2016 at 2:22 AM, Larry Colen <[email protected]> wrote:

> In simple terms, the definition of infinity focus would be to have parallel
> incident light rays. Divergent rays from a point source are closer than
> infinity. One could probably set up a calibration unit with parallel light
> rays using conventional optics. I suspect that a low powered laser would
> also work.

In case it helps your googling, what you're trying to achieve is
"collimation" of light.

When I taught astronomical instrumentation lab, we achieved this on
the bench with a white light shining onto a spatial filter (basically
a pinhole), with the spatial filter located in the focal plane of a
Nikon 200/2.8 or 300/4 lens (which was "locked down" to infinity
focus). The light coming out the front of the lens was (reasonably)
collimated. If you picture parallel rays coming from a star on the sky
and being focused to a point on the sensor, it's exactly that, except
the light is moving in reverse.

We then sent the collimated light into rudimentary lenses (e.g.
achromatic doublets) and viewed the spots produced, to compare the
actual aberrations with predicted spot diagrams from optical design
software.

The big caveats are:
1) Your collimating lens has to be focused at infinity and the right
distance from the point source, or the light that comes out isn't
collimated (it will "look" closer or farther than infinity). So this
just shifts the problem of finding infinity focus to another optic!
You may be able to get calibrated collimating lenses, making it a
mechanical positioning problem.
2) At least if you're trying to evaluate aberrations of the optics
under test, the collimating lens has to be optically superior to the
test lens, or its own aberrations will come into play. In our case,
with good Nikon lenses against doublets or triplets, this was the
case.

-- 
PDML Pentax-Discuss Mail List
[email protected]
http://pdml.net/mailman/listinfo/pdml_pdml.net
to UNSUBSCRIBE from the PDML, please visit the link directly above and follow 
the directions.

Reply via email to