RE: [ATM] Foucault and diffraction

["Koehler, Steve" <Steve_Koehler@securecomputing.com>]

Vlad,

> Thank you, an image is worth a thousand words. What I
> realized after my initial post is that what we have in the focal zone
> is not a point-source pattern, but the source of light re-imaged in its
> original size. Since the source of light normally is not much smaller
> than 0.1mm, either for a slit or (especially) a  pinhole, it is still
> nearly ten times the diameter of the light cylinder of a point-source
> form.
> This means that the actual form is also nearly ten times wider,
> making the bright cylinder of the highest intensity even more pronounced.

That's a very good point.  My slice patterns are for a point source.  If you
aren't aware of it, the way all of us (I believe) are simulating a non-point
source in the Foucault and related tests is to do the diffraction calculations
using a point source, but soften the edge of the tool (e.g., knife) by an amount
depending on the source size.  I have yet to be convinced that this always
works, but it appears to do very well, indeed, when comparing with actual
images.

I believe I could use the formulas for incoherent illumination to switch the
longitudinal slice patterns over to source with actual extent.  This would be
more correct than what I have done.  If I do this, I just won't be able to
reverse the process to see what the shadows look like when a knife is
introduced.

> I was planning on looking at the recommended programs,
> which may help me to make it a bit beyond gross generalizations.
> Also, maybe to understand better what is happening when spot-testing.
> It is interesting, isn't it: like opening up a toy to find out what's
> inside.
> It always results in something: even just a pile of parts that I don't
> know
> what to do with... :)

If you're programming language literate, I would be happy to share my tools with
you.  I have not completed documentation on them, yet.  The advantages of my
tool set are that it is a superset of the functionality of Diffract.exe and
Aberrator, that it is often easy to modify for new capabililties, and it is
scriptable so repetitive operations can be done by writing a program to do them.
Please ask, if you can't get the job done with the other available tools.

-- Steve Koehler
   steve_koehler@securecomputing.com

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