Tip-tilt adaptive optics needed

Moderators: Sam, Sean.Wang

Benni
Posts: 47
Joined: Sat Mar 11, 2017 9:52 pm

Re: Tip-tilt adaptive optics needed

Post by Benni » Wed Jan 16, 2019 3:07 pm

I guess an active optics would only make sense in such a configuration.
One reason is this:

Many telescopes, for example Newtonians with flatteners, need a backfocus of 55mm (i.e that of a typical dslr with t2).
This has practical reasons. One wants to put the chip as close to the secondary as possible in order to avoid vignetting.

An active optics, if one puts it in front of a guider and a filter wheel simply takes too much space. It therefore is only good for some refractors, Schmidt Cassegrains or Richey Chretiens.

But a chip sitting on piezo elements that can move one or two milimeter sideways, with some guiding and seeing correction, does not take much backfocus.

But that would work only if there is a separate guiding chip on the same board.

And with filters, a separate guiding chip next to the main chip works only if the filters are not covering the guiding chip...

And that then works only if the filters are sitting very close in front of the main chip.

I think one can build a filter wheel into the camera such that the wheel does not obstruct the guiding chip.

Maybe one can put some baffles around the main chip and the guiding chip in order to protect the two chips from getting stray light...

Such a camera, (integrated filter wheel, self guiding, and guiding chip is not obstructed by filters) is certainly possible, I think...

This would work with 55 mm backfocus on newtonians or even less.... and it would be considerably lightweight. There would no need for heavy glass and metal like in a separate active optics unit or with an off axis guider...
and the piezo elements could correct seeing at 100 hz. Perhaps the camera may have some hardware that it can make some guiding on its own without the need to download every image to a pc...

Then one could almost go to the optical limit of the telescope...
Almost because seeing also contains blow ups and these can not be corrected by piezo elements, but they are considerably rare...

User avatar
Sean.Wang
Posts: 713
Joined: Mon Apr 24, 2017 10:08 am

Re: Tip-tilt adaptive optics needed

Post by Sean.Wang » Thu Jan 17, 2019 2:29 am

Benni wrote:
Wed Jan 16, 2019 3:07 pm
I guess an active optics would only make sense in such a configuration.
One reason is this:

Many telescopes, for example Newtonians with flatteners, need a backfocus of 55mm (i.e that of a typical dslr with t2).
This has practical reasons. One wants to put the chip as close to the secondary as possible in order to avoid vignetting.

An active optics, if one puts it in front of a guider and a filter wheel simply takes too much space. It therefore is only good for some refractors, Schmidt Cassegrains or Richey Chretiens.

But a chip sitting on piezo elements that can move one or two milimeter sideways, with some guiding and seeing correction, does not take much backfocus.

But that would work only if there is a separate guiding chip on the same board.

And with filters, a separate guiding chip next to the main chip works only if the filters are not covering the guiding chip...

And that then works only if the filters are sitting very close in front of the main chip.

I think one can build a filter wheel into the camera such that the wheel does not obstruct the guiding chip.

Maybe one can put some baffles around the main chip and the guiding chip in order to protect the two chips from getting stray light...

Such a camera, (integrated filter wheel, self guiding, and guiding chip is not obstructed by filters) is certainly possible, I think...

This would work with 55 mm backfocus on newtonians or even less.... and it would be considerably lightweight. There would no need for heavy glass and metal like in a separate active optics unit or with an off axis guider...
and the piezo elements could correct seeing at 100 hz. Perhaps the camera may have some hardware that it can make some guiding on its own without the need to download every image to a pc...

Then one could almost go to the optical limit of the telescope...
Almost because seeing also contains blow ups and these can not be corrected by piezo elements, but they are considerably rare...
Thank you so much, your suggestion is very valuable to us.
ZWO | Marketing Consultant
Location:lon=120.6 lat=31.3
SuZhou China

motto:
Don't face one's fate, can't be a real man;
Don't know manners, can't be gentle;
Don't know meaning behind the words, can't understand man's heart.

Benni
Posts: 47
Joined: Sat Mar 11, 2017 9:52 pm

Re: Tip-tilt adaptive optics needed

Post by Benni » Sat Jan 19, 2019 1:00 am

Well,
if you are developing it and make much money from my proposed camera design, don't forget to send me a working camera...

Would really like to work with such a thing....

Currently, I only have an asi 1600 which is also very good....

User avatar
Sean.Wang
Posts: 713
Joined: Mon Apr 24, 2017 10:08 am

Re: Tip-tilt adaptive optics needed

Post by Sean.Wang » Thu Jan 24, 2019 9:09 am

Benni wrote:
Sat Jan 19, 2019 1:00 am
Well,
if you are developing it and make much money from my proposed camera design, don't forget to send me a working camera...

Would really like to work with such a thing....

Currently, I only have an asi 1600 which is also very good....
oh, yes, no problem. we didn't do that now, your idea is great!
ZWO | Marketing Consultant
Location:lon=120.6 lat=31.3
SuZhou China

motto:
Don't face one's fate, can't be a real man;
Don't know manners, can't be gentle;
Don't know meaning behind the words, can't understand man's heart.

Benni
Posts: 47
Joined: Sat Mar 11, 2017 9:52 pm

Re: Tip-tilt adaptive optics needed

Post by Benni » Thu May 09, 2019 3:47 am

Thanks for the offer.

I currently use an starlight express adaptive optics. I could put the camera on and compare how the two adaptive optics work.
I currently use a celestron edge 800 and a Ts 65 flat field APO. Adaptive optics would benefit more from higher focal lengths, like on my edge.

However, I live in munich with much light pollution and I don't have clear skies very often.
And my edge sometimes needs collimation...

Furthermore, I only do, i compared to other astrobin results, only average quality photos of dso's,
So there are certainly better photographers around who have sites with less light pollution who could test this camera...

But it would be interesting if one can do away with this heavy sx ao...

For the design:

It may be better if only the main chip is moved on piezo crystals and the movement of the guiding chip is done in software. Then it is cheaper. And one can put the main chip in some kind of housing together with the filter wheel, to prevent straylight, whereas the guiding chip sits directly next to this structure. I do not know if the guiding chip would benefit much from cooling. If one does not see additional stars, then one can save the cooling for the guiding chip too and simply place it next to the long end of the main chip plus wheel.
I guess that should be doable with an M48 opening. But I am no designer...

Benni
Posts: 47
Joined: Sat Mar 11, 2017 9:52 pm

Re: Tip-tilt adaptive optics needed

Post by Benni » Sun Jul 28, 2019 7:05 pm

I want to note that there are now cameras with sensor image stabilisation:

https://www.dpreview.com/news/779913205 ... -hybrid-af

Now you need to reduce this 5 axis thing to 2 axes.
Then one can put a filterwheel directly on top of that sensor,
Finally make a housing for that wheel to prevent light going in, and put a guiding chip directly next to that wheel housing, that both guiding and main sensor can get into focus.
The guiding chip does not even need to move. I guess that can be done in software. Important is that the main chip can move.

And that the sensor movement must be fast. At best 100hz of course. Coupled with a high resolution sensor, one gets basically a scientific device for which seing is no problem.

Benni
Posts: 47
Joined: Sat Mar 11, 2017 9:52 pm

Re: Tip-tilt adaptive optics needed

Post by Benni » Sat Aug 03, 2019 9:00 pm

Perhaps one could even leave it with 5 axes movement and use the other axes as automatic tilters, perhaps with some software that computes the optimal directional tilt of the sensor for a telescope and then the rest of the axes are used for guiding, corrected by a fixed chip that sits nearby, outside the housing of the main sensor and filter wheel.

But i guess the cooling is then difficult. the tilter must perhaps move the entire unit of sensor and peltier.

if there are problems to get the guiding sensor next to the wheel housing, i guess one can place a small prism on the top of the internal filter wheel housing. the prism can direct the light somewhere else where the guiding sensor sits in the correct distance.

RegMaster
Posts: 2
Joined: Fri Apr 20, 2018 11:30 am

Re: Tip-tilt adaptive optics needed

Post by RegMaster » Mon Sep 23, 2019 3:48 pm

In my opinion the historical problem with the active optics of the "amateur" world has been given by several factors.

The first is that I consider it a mistake to use an active optics as a autoguide system. I think that the mount should have its own "on axis" guidance system, and use the active optics only to solve seeing problems and high frequency inaccuracies. The active optics system would need another camera independent of the mount autoguide camera, for example, an ASI174 mini (medium chip size), and that the corrections of the active optics are a maximum of 3 arcseconds and high frequency corrections 1/10 sec, approx.

I think that these systems are ideal for high resolution configurations 1 arcosecond per pixel or less, like the configurations with large cassegrains or RC. I don't see useful to mount this system in newtons or refractors (professional astro photographers know this).

Finally comment that another problem has been given by latencies, olds computers and calculation capacity for 10 fps or more, Today with usb 3.0 interface of low latency and powerful processors, there should be no problem to operate (with ROI) 10fps or more with low latency and make really effective active optics in the amateur world.

I think too that de active optics system must be installed between telescope and filterwheel.

Greetings.

Benni
Posts: 47
Joined: Sat Mar 11, 2017 9:52 pm

Re: Tip-tilt adaptive optics needed

Post by Benni » Fri Sep 27, 2019 5:27 pm

RegMaster wrote:
Mon Sep 23, 2019 3:48 pm
The first is that I consider it a mistake to use an active optics as a autoguide system. I think that the mount should have its own "on axis" guidance system, and use the active optics only to solve seeing problems and high frequency inaccuracies. The active optics system would need another camera independent of the mount autoguide camera, for example, an ASI174 mini (medium chip size), and that the corrections of the active optics are a maximum of 3 arcseconds and high frequency corrections 1/10 sec, approx.

I think that these systems are ideal for high resolution configurations 1 arcosecond per pixel or less, like the configurations with large cassegrains or RC. I don't see useful to mount this system in newtons or refractors (professional astro photographers know this).

Finally comment that another problem has been given by latencies, olds computers and calculation capacity for 10 fps or more, Today with usb 3.0 interface of low latency and powerful processors, there should be no problem to operate (with ROI) 10fps or more with low latency and make really effective active optics in the amateur world.

I think too that de active optics system must be installed between telescope and filterwheel.

Greetings.
It appears you do not have used an ao system yourself.
Usually, the ao can command both itself and the mount. So if the deviation becomes large, it would issue a mount movement, while otherwise, it corrects with the ao.

The problem with the usual solutions in market is that they require backfocus and that they are expensive.

For example, the sx ao from starlight xpress takes 47mm backfocus plus 6mm for the adapters. If you have a telescope with just 55mm backfocus after the corrector, the sx ao simply would not fit.

Therefore, my suggestion to implement the ao mechanism via piezo elements that move the ccd and peltier assembly.

As for what are the benefits:

Well basically it would benefit any telescope. If the ao and guiding mechanism is slow, there is only a real benefit if you have a very poor or old mount.

On the other hand, If the mount is good, and if the ao is good, then the ao can reduce seing a bit.

Here, somebody has made comparisions for images with and without ao:

https://astroanarchy.blogspot.com/2007/ ... light.html

I guess I can confirm this.

This effect does not only happen for very large rc's or sc telescopes but also at rather moderate focal lengths.

The stars that are taken with the ao simply do not have that large halos around them.

The question is, whether that effect is important for you to pay 1000 euros for the ao....

This is why i suggest to put the ao into the cameras themselves.

If one simply uses the 5 axis sensor stabilisation systems, then this may have other uses, for example, together with software, one can make an automatic tilt corrector. The software can take images, and automatically adjust the sensor tilt. And with the other axes, it can guide at fast frequencies.

If the guiding chip is placed next to the sensor, and has a separate, unfiltered lightpath from the main imaging sensor over which the filterwheel rotates, the entire camera then does not need additional things like off axis guiders... it simply needs less additional stuff...

RegMaster
Posts: 2
Joined: Fri Apr 20, 2018 11:30 am

Re: Tip-tilt adaptive optics needed

Post by RegMaster » Mon Sep 30, 2019 9:08 am

Benni wrote:
Fri Sep 27, 2019 5:27 pm
RegMaster wrote:
Mon Sep 23, 2019 3:48 pm
The first is that I consider it a mistake to use an active optics as a autoguide system. I think that the mount should have its own "on axis" guidance system, and use the active optics only to solve seeing problems and high frequency inaccuracies. The active optics system would need another camera independent of the mount autoguide camera, for example, an ASI174 mini (medium chip size), and that the corrections of the active optics are a maximum of 3 arcseconds and high frequency corrections 1/10 sec, approx.

I think that these systems are ideal for high resolution configurations 1 arcosecond per pixel or less, like the configurations with large cassegrains or RC. I don't see useful to mount this system in newtons or refractors (professional astro photographers know this).

Finally comment that another problem has been given by latencies, olds computers and calculation capacity for 10 fps or more, Today with usb 3.0 interface of low latency and powerful processors, there should be no problem to operate (with ROI) 10fps or more with low latency and make really effective active optics in the amateur world.

I think too that de active optics system must be installed between telescope and filterwheel.

Greetings.
It appears you do not have used an ao system yourself.
Usually, the ao can command both itself and the mount. So if the deviation becomes large, it would issue a mount movement, while otherwise, it corrects with the ao.

The problem with the usual solutions in market is that they require backfocus and that they are expensive.

For example, the sx ao from starlight xpress takes 47mm backfocus plus 6mm for the adapters. If you have a telescope with just 55mm backfocus after the corrector, the sx ao simply would not fit.

Therefore, my suggestion to implement the ao mechanism via piezo elements that move the ccd and peltier assembly.

As for what are the benefits:

Well basically it would benefit any telescope. If the ao and guiding mechanism is slow, there is only a real benefit if you have a very poor or old mount.

On the other hand, If the mount is good, and if the ao is good, then the ao can reduce seing a bit.

Here, somebody has made comparisions for images with and without ao:

https://astroanarchy.blogspot.com/2007/ ... light.html

I guess I can confirm this.

This effect does not only happen for very large rc's or sc telescopes but also at rather moderate focal lengths.

The stars that are taken with the ao simply do not have that large halos around them.

The question is, whether that effect is important for you to pay 1000 euros for the ao....

This is why i suggest to put the ao into the cameras themselves.

If one simply uses the 5 axis sensor stabilisation systems, then this may have other uses, for example, together with software, one can make an automatic tilt corrector. The software can take images, and automatically adjust the sensor tilt. And with the other axes, it can guide at fast frequencies.

If the guiding chip is placed next to the sensor, and has a separate, unfiltered lightpath from the main imaging sensor over which the filterwheel rotates, the entire camera then does not need additional things like off axis guiders... it simply needs less additional stuff...
No, I haven't used active optics, but I've seen her use some of my friends
And I'm just tired of seeing examples like the one you gave of the bubble nebula.

What good is it to use active optics if I don't solve other problems before, such as poor focus, drift, oversampling, misguided, etc...??? you give me an example with a 12" telescope that is supposed to be able to take high resolution photos. It says that the photo is made with a QHY8, with a pixel of 7.8u, if we count that the Meade of 12" uses focal reducer 0.67x... we have that the resolution of the photo would be 0.8arcs/pixel + active optics working.

In this photo of a colleague (AIP): https://aipastroimaging.com/ngc7635-hoo/

You can see the bubble much sharper, without errors in guidance or focus, made from Spain, at 800m over sea level, arox. ... without using active optics or any other corrective method, with a 4.5" telescope (TS115) and Atik 460, leaving us with a resolution of 1.47 arcs/pix.

How is it possible that a photo of 0.8 arcs/pix + AO is more blurred than a photo of 1.47 arcs/pix without AO?

Easy, because the AO is not working as it should, trying to correct misguided, focus problems, etc..

I know that the Starlight AO model for example can operate corrections to the mount if the drift is large, but I think it is totally counterproductive. The OA should only focus on correcting the medium frequency seeing (10hz).

Post Reply