Questions SBIG AO-7 and Other Adaptive Optics - Long

Hello everybody,

I'm contemplating purchasing a SBIG AO-7 as being the only so called
adaptive optics device currently on the market . I'd like to know :

1 - are there any other alternatives ? I found some references to a
discontinued Stellar Products device and a higher order correction device ,
but their website doesn't show if this is a current product or just a future
project . I'd appreciate any advice, personal knowledge or experience with
these .

2 - the SBIG AO-7 is marketed by SBIG as a tip/tilt adaptive optics
device. Allegedly it corrects for mount fast tracking errors plus
atmospheric turbulence induced tip/tilt.
However, it works only with the SBIG dual head cameras . It uses the
guiding head to generate the tip/tilt error signal, and hereby is the crux
of the problem. Given the considerable off axis position of the second CCD
(the guiding TC-237 CCD) , the separation between the (larger) imaging CCD
center and the guiding CCD is too large and the guiding CCD is so to speak
out of the isoplanatic angle . In other words, the guiding CCD is not seeing
the same turbulence as the imaging CCD , but a different tip/tilt motion due
to its large off axis distance. This is true for any reasonable image
sampling ,unless using an extremely (unusually) long focal length
instrument, in which case the image sampling is much too fine , exceeding by
far the scope resolution, and narrowing the FOV to almost nothing , reducing
image birghtness and increasing exposure time tremendously .

3 - based on well known very reputable references related to atmospheric
turbulence (and on my own measurements with a DIMM setup ) , the incoming
wavefront tip/tilt needs a sampling frequency in the hundreds of Hertz. I
can post links to these papers stemming from research and researchers at the
world's largest telescopes . The AO-7 is much slower, although how slow
exactly is debatable due to the imprecise and fuzzy specs on the SBIG
website .
Here's exactly what they say "Utilizing the second guiding CCD as an
imaging sensor the guide star's position is read out at rates up to 40 times
a second, and the tip-tilt mirror adjusted to hold the star on the
designated pixel for the length of the imaging CCD's exposure. The tip-tilt
mirror moves and settles to within 20% of the commanded position in a mere
10 milliseconds".
The above quote is not easy to understand , as it seems to be either the
product of marketing/management (as opposed to engineering or scientific
truth) , or the product of marrying 2 devices that were shoehorned into a
product without actually being suitable for it . Let me explain. If the
guide star position is read out 40 times a second, it means the guiding
camera has a frame rate of 40Hz. The frame duration is 25 ms (milliseconds)
..
If the guide camera is read by the Windows PC software, this operating
system introduces non trivial delays as well, at least adding a few
milliseconds , let's say 5ms to pick a number (which is in the usual range ,
depending largely on the PC speed, type, operating system , installed
software, shoe size etc) . The overall delay from the moment a frame started
being exposed to the moment the PC calculated the tip/tilt correction will
be 25ms+5ms=30ms . To satisfy Nyquist, the signal bandwidth must be less
than half, meaning a signal period of 60ms and correction rate of 15Hz .
Then SBIG adds the puzzling part, claiming the tip/tilt mirror settles to
within 20% in a mere 10 milliseconds. They must be claiming here strictly
the mirror +actuators settling time and NOT the overall system settling time
after a wavefront tip/tilt is introduced . How much faster the mirror moves
is irrelevant, since the correction will take place always at the 15Hz rate
, not at the implied higher frequency due to the 10ms settling time .
Where does that leave the system? Lower on the webpage, there's a claim
that "This remarkable system promises to have a profound effect on CCD
imaging by reducing the atmospheric turbulence, wind induced vibrations, and
eliminating the remaining periodic errors in most telescope drives." It is
true that most scopes will have or can be reasonably forced into having wind
induced vibrations in the under 15Hz range, so the AO-7 corrects no doubt
for wind gust induced vibrations. It is also true that most mounts will
exhibit slow changing PE , and a correction bandwidth of 15Hz will take
care of that. Atmospheric turbulence on the other hand, at less than 15Hz
correction, no way . That is the pure marketing hype, and that's what gives
the device the name of AO .

To recap, AO-7 is NOT AO because:
-the guiding ccd is too far off axis and images a different atmospheric
turbulence pattern than the main imaging ccd . As a matter of fact, it
probably corrects for a totally different atmospheric turbulence than the
main CCD is seeing .
-the sample rate is too low by an order of magnitude to correct for
atmospheric turbulence .
I'm sure the AO-7 is a great device, I'm sure it helps with mount
vibrations , wind induced vibrations etc, improves images etc, but is NOT an
AO device.
Please correct me if I'm wrong, since I'd like to buy an AO device . If
it's not going to be the SBIG AO7, then there's nothing else on the market .
Help , please.

Thanks,
Matt Tudor

On Wed, 27 Oct 2004 21:04:29 -0400, "matt" wrote:

1 - are there any other alternatives ? I found some references to a
discontinued Stellar Products device and a higher order correction device ,
but their website doesn't show if this is a current product or just a future
project . I'd appreciate any advice, personal knowledge or experience with
these .

Not unless you make them yourself.


2 - the SBIG AO-7 is marketed by SBIG as a tip/tilt adaptive optics
device. Allegedly it corrects for mount fast tracking errors plus
atmospheric turbulence induced tip/tilt...

Your analysis is essentially correct. The AO7 does not correct for atmospheric
effects in any significant way. It corrects for guiding and tracking errors in
the mount and for some vibration in the mechanical system. I believe it is fair
to call such a device an active optical element, however, since it does modify
the final image by altering the optical path.

_________________________________________________

Chris L Peterson
Cloudbait Observatory
http://www.cloudbait.com

"matt" wrote

I'm contemplating purchasing a SBIG AO-7 as being the only so called
adaptive optics device currently on the market ............


......... Please correct me if I'm wrong, since I'd like to buy an AO
device . If
it's not going to be the SBIG AO7, then there's nothing else on the market
..
Help , please.

Matt,

As far as I know there is nothing else currently for sale other than the
AO-7, and soon its "big brother" for the SBIG STL cameras. SBIG also
mentioned something about possibly selling a cheaper slower "tracker"
version.

Most people contend that the AO-7 does next to nothing for seeing but it
is great for improving guiding for problems like mount glitches and shifting
optics. As for your point about the guider-chip being off-axis: At one time
about a year ago or so SBIG had some info on their website noting that the
AO-7 would work much better on correcting seeing if it could guide on a star
that was near the center of the imaging chip. They claimed to be working on
an add-on to do this which I believe was (is?) a beam-splitter that would
send IR over to the guide-chip. However SBIG's efforts must not have worked
out too well because I've not heard anything on it for a while. Maybe it
just went on the back burner.

When I called SBIG and talked to them they basically said that the AO-7
works to improve seeing mostly on large telescopes, which is OK for me since
I mostly use a 20-inch F/8 RC Cass. They said that there would be little
improvement of 'seeing' on my 10-inch but it should help on the 20. In the
manual for the AO-7 there is a section that says something like 'Take a very
short sub-second exposure of a bright star. The AO-7 will not be able to
improve the FWHM beyond what you measure in such short exposures".

I've found that my AO-7 improves FWHM about 20% to 50% or more on the
three telescopes I've tried it on. On a 60 second exposure with my ST-9/AO-7
on the 20" F/8, the FWHM will go from say 3.5 to 2.5 when using the AO. I'm
not sure what it's correcting, but it works. On the other hand, I've found
my AO-7 to be more delicate than the SBIG cameras and more prone to needing
repairs. I would not want to drop it, and SBIG did have a certain number of
them not survive shipping (mine being one - It was DOA and I had to send it
back for adjustments). You also 'pay' for your better resolution with added
set-up time and complexity when imaging, plus it's just another gizmo to
take along to the observatory or out in the field. It also makes the camera
heavier and more difficult to mount on a telescope. I now only use the AO-7
about half the time when I'm really concerned about improving resolution. It
has also helped in the field on those nights I've been too lazy to do a full
drift polar alignment.

You might find interesting the following page of info offered by the guys
who 'invented' the AO-7 and sold it to SBIG:
http://voltaire.csun.edu/ao.html You can also download the manual at the
SBIG website and perhaps gain some feeling on how it works from the section
of the CCDsoft and CCDops manuals discussing it.

George Normandin

I have heard that Starlight Xpress are currently developing an "Active Optics"
equivalent to the AO7. The SX product will be a straight through configuration
(tip/tilt optical window) and will essentially correct the same problems as the
AO7 - that is high speed guiding and image shift corrections . The preliminary
specs are 10ms response, +/-75 microns image shift, and the image shift
relationship to the input data is well defined, This should allow 'open loop'
guiding with a guide telescope as well as conventional closed loop control with
the guide device behind the AO unit.

I hear it is due for release at the end of the year.

Mark