Re: [ATM] holographic null test on round robin mirror C

["matt" <mariusrf@bellsouth.net>]

Dale, please see my inline answers:

-----Original Message-----
From: Dale Eason <atmpob@yahoo.com>
To: matt <mariusrf@bellsouth.net>
Date: Thursday, February 03, 2005 6:46 PM
Subject: Re: [ATM] holographic null test on round robin mirror C


>The hartman test also requires anaysis software.  The
>hologrphic test is a null test.

yes, the Hartman test requires analysis software , as does interferometry .
The software is similar in its Zernike fitting to your interferometry
software. I think you could actually reuse it, or at least theoretically
write a Hartman capture section that would grab the image , find the
subaperture theoretical centroids, compare with measured centroids, then
derive local slopes and from there you're getting into the common side with
Zernike fitting and everything that follows .

>
>Jim Burrows is working on and has published his 2D
>hartman test.  I used it.  It has two complications,
>1. Making the mask is tedious and the mask must be
>flush against the surface in his version.  That means
>you cannot use it on coated mirrors if you value the
>surface.  The analysis requires a good deal of your
>time button pressing  (about 20 minutes) and you have
>to decide when to stop.
>

I agree. However, it doesn't need to be like that. Actually, no other
Hartman analysis method has the requirement for square holes mask, or for
the mask next to the optics, both being show stoppers for me .

>2. You have to use a tiny pinhole light source and
>carefully measure three distances.  I found the setup
>very hard.
>

I have a couple of experimental wavefront sensors that I made, and the
calibration procedure involves first centering the mask or lenslet array on
the CCD, then centering a pinhole at 2 distances , like 1m and 2m . I made a
simple calibration fixture in a small diameter adequately long tube  . The
software will then calculate everything with no button pushing .

>Its advantage is that it is doable without a lot of
>expense.  Only the cost of the web camera and a good
>lens.  Some will already have an usable camera.
>Interferometers are more expensive.


its main advantages are:
- with a mask or lenslet array of 20x20 subapertures for example, which is
very modest (professional observatories use larger numbers, 100x100 for
example) , you get 400 surface points sampling. That is comparable to
interferometry , say 10 fringes with 40 points per fringe.
-it samples the whole mirror area not just one diameter
-it is insensitive to vibrations. No more home geophone instead of mirror
testing.
-it is a lot less sensitive to local seeing than interferometry . Local
seeing can be easily integrated with longer exposure times .
-no coherence length requirements for light source , good image contrast ,
can stack images .

 best regards,
matt tudor

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