Re: [ATM] Re: Any Amateur Adaptive or Active Optics

["Greg Jones" <gregj888@comcast.net>]

Well, let's see...
1) 5 counts tip and tilt... add focus and 2 orders of astigmatism.

2) Site (Ro) and imaging wavelength (Ro is related to wavelength).  Just a
guess right now.

3) CCD for the sensor, yes 4 quads.  The lenses have wedge prisms behind
them to allow alignment along the bottom of the CCD (quicker readout).  Mag
limit for 100Hz loop is hoped to be 10 -12, but will depend on sensor QE.
Have a TI-211 right now, keep hoping for something better.  The big guys use
APD proton counters at $5k per channel, I need 16...  I have a CCD.
Actually if there was a chance to get them I would use a curvature sensor
and would only need 4.

4) The hope is a transformation martix will do the trick.  With my meager 4
"zones" I think I can make it work, but would be difficult for a high order
system.  In that case you would model the wave front then calc. the
correction.  For those scratching their heads- the sensor works in tip/tilt
the corrector in focus.  Much like the old X/Y graph vs polar.  Conversion
is thought to be similar.  As far as I know they no longer pair
sensor/actuators, there are several papers that explain why.

5) No, hope to, but is less an issue for me.  I'm not sure I know what a
DIMM is.  Anyway, there may be a difference between what you want to do and
what you can do.  I have a 20" mirror and 4 subapertures (~10").  You don't
have much light below that, so that's one issue.  Next is the Ro number,
which conceptually is the max size of a defraction limited scope.  At
sea-level and green light Ro might be real small an inch, maybe less, maybe
more; at that point you're just out of luck.  If Ro is 4 inches, a 12" scope
will still likely be better than 1/2 wave due to higher orders (call it
boiling) but the 1 arc-sec image may move 2-3 arc-sec due to tip/tilt.
Tip/tilt/focus would get you a long way on a 12" scope and for now about all
you can get (IMHO).

Also note that for AO to help you have to be able to resolve at the
defraction limit across 2 pixels.  So again, roughly speaking (Jones' rules
:-), the F number needs to be 2x the pixel size in microns.  My 24 micron
pixels means operating around f50...  If your CCD has 8um pixels you'll need
f16 or so.

BTW I have a good stash of PZT material if you need some.

Do you have a web site?  What is your system like?

Greg Jones

----- Original Message ----- 
From: "matt" <mariusrf@bellsouth.net>
To: "Greg Jones" <gregj888@comcast.net>
Sent: Wednesday, June 23, 2004 11:04 PM
Subject: Re: [ATM] Re: Any Amateur Adaptive or Active Optics


> Hi Greg,
>
> Thanks for the reply . I'm working on an adaptive optics system too but
> you're far ahead of me .
>
> I do have a few questions :
>
> 1- how did you get 5 orders of correction with only 4 subapertures ? Most
> references I've seen are quoting more along the lines of correction order
> being around 60% of number of subapertures, after tip/tilt is eliminated .
>
> 2-what is the limiting factor for 5-8 arcsec for correction radius? Lower
> order corrections should be valid for a larger angle .
>
> 3-are you using 4 quad cells ? From the image it appears that you have a
> Hartmann mask with 4 quads , although the website says Shack Hartmann .
What
> magnitude do you need the guidestar for full bandwidth correction ?
>
> 4-how did you calculate the corrections for the bimporh mirror? most
> literature references quote Shack Hartmann paired with discrete actuators
on
> the back of a thin mirror, and curvature wavefront sensors with bimorph
> mirrors .
>
> 5-with an average isoplanatic patch of 4" , fast tip/tilt is not enough
for
> a 12" typically, unless you have some unusual seeing there . Have you
> measured your local turbulence in any way ? I have a DIMM setup and am
> analyzing the first batches of data from it , it is by far the easiest way
> to quantify seeing for me.
>
> thanks,
> matt tudor

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