Adaptive Optics versus low PE mount

[Kal (Andrew)]

I'm trying to find out just how much adapive optics will correct for mount PE. I know that adaptive optics units like the AO-7 and AO-L will remove alot of the effect of periodic error, moreso than autoguiding can, but I'm wondering just how much. Specifically, I'm wanting to know of what sort of peak to peak PE I could expect on an LX200 on a wedge with an adaptive optics unit (such as the SBIG or starlight express one) in place, compared to autoguiding without AO.

Does anyone have any experience with this, or have you seen any good reviews/articles?

[[1ponders] (Paul)]

Lot's of good info on LX200 here (not sure if meade have upgraded the mount in the interum for the "R" ) but the PE info is about half way down the page.
http://www.richweb.f9.co.uk/astro/notes-2004-12-07.htm

I doubt that the AO would be able to deal with 50 arcsec on it's own. Probably after PE training it would.

[g__day (Matthew)]

Cloudy nights did a review (my journey to astrophotography over several years - parts 1 and 2) about 2-3 months ago. In part one it had an interesting segment on the persons reliance on adaptive optics and it actually sent him backwards - disguising other problems and limitations in his set up and methods - after spending US $1,500 for the AO-7 I believe.

It's well worth a read to include in your thinking.

http://www.cloudynights.com/item.php?item_id=1647

excerpt -

The first wonderful piece of advice I got (and ignored of course) was to not use the AO-7 and simply use the camera and guider directly into the CGE. But I thought, “My CGE needs the AO-7 to get good tracking” and used it right from the beginning. The trouble was for the first few months my tracking was 50/50. Some nights it worked flawlessly and other nights it couldn’t keep the guidestar for more that a few minutes. Then I found and read Mike Dodd’s excellent article that explained auto-guiding in detail, located on the Yahoo Group “CGE-Uncensored”. From then on I had no issues with guiding and it worked flawlessly every time. One major issue solved but the AO-7 still was having a detrimental impact on my images but I didn’t know it yet.

...

The biggest revelation came in December, a whole 8 months after starting this endeavor. In one of the CN threads a link was given to Richard Bennion’s presentation at an imaging conference. I watched the 45 minute video and all was revealed. I learned that my collimation wasn’t good enough, my star distortion was an issue, my focus wasn’t quite good enough, and how important all of these pieces are to the final image. I removed the AO-7 to improve the overall image quality because the focus point with it was near the extreme back and without it was much closer to the middle. I spent days playing around with collimation trying to judge exactly why my stars were slightly oval and got them spot-on circular with the brightest point at the center pixel. I was finally able to get FocusMax to work extremely well and understood what HFD means and what my typical HFD values should be on good nights and bad nights.

[Bassnut (Fred)]

Andrew

There are lots of issues useing an A07 that may cause grief despite the bonus eliminating PE effects.

Having said that, when its working well, its a wonder to behold !, an awesome machine.

To use it, you need an SBIG cam (internal guiding).

As they can work up at to 40hz, they can hide a lot of bad PE at that frequency, not sure about 50 arcsecs, but my G11 mount at the time I tried had at least 15-20 arcsec PE (just out of the box, no tweaking, and overloaded to hell) and it creamed that. It will also nudge the mount as in normal guiding if the guide star gets too close to the edge.

The downside is of course you need a VERY bright star to get anything like 40hz, that is exteremely hard to find. I found with a clear filter you were lucky to get much faster than 0.5 secs (well, without constantly rotating the cam to find a better one, which is a real pain without an instrument rotator).

Operating through filters, I found I was back to 1 to 2 sec guiding exposures too often, which gives you no advantage (or dealing with PE) over normal guiding.

If your really hard core, or simply cant get PE down on your mount to get really tight subs, then an A07 really is the bomb, no dought, but its a mongrel to use, and you have to be very patient to it set up.

I gave up, cant hack it. I just kept trying to reduce PE, Pempro is very good to generate PEC, consider that too.

Cheers
Fred

[Kal (Andrew)]

Some good material there and some interesting points for me to take on board.

As a note I don't even have a wedge yet, but I will be getting one soon and I am just looking at options for down the track when I move on from my DSI. From what I previously read it looked like adaptive optics could give me excellent results, but what g_day posted is quite valid - perhaps I am looking at this wrong by thinking a wedge mounted LX200 needs AO to obtain good tracking.

[Bassnut (Fred)]

Andrew

Get a wedge, get PE right down with PEC (really, pempro will help a lot there) and see how it goes. I see many great pics from well tuned LX200Rs (a FR also would help, say the meade F6.3).An A07 and SBIG cam is a big jump, you would be better off with a DSLR 1st IMO, and an external guide scope and webby for guiding, at a fraction of the price.

Eventually, you would have to upgrade the mount if your serious about astrophotography (and PE ;-), its the single most important item in the whole rig. And after that it starts getting really, really expensive, dont even bother if you dont have $20k minimum in the bank ;-).

Cheers
Fred

[g__day (Matthew)]

After seeing Fred's latest shots on the above link I am simply amazed at how good you can do with a guided G11 and an S-Big + alot of talent.

What do you use to cut light pollution on your longer shots Fred? I'm in North RYde and my shots past 8 minutes (using a Canon 400D are flooded with too much ambient light pollution)?

[Kal (Andrew)]

Thanks for the direct feedback from someone that has used it before Fred.

Apart from having difficulty finding suitable guide stars to get the short subs required for the AO to work to the point where it provides a benefit, are there any other hassles with using the unit that made it difficult to work with?

[Bassnut (Fred)]

Well, my guiding aint perfect either, I just make it so you cant tell . To get over ordinary guiding/bad seeing/light pollution, the secret weapons are lots of data, as much as you can get (although, it quickly gets silly, deminishing returns) and CS, I spend lots of time with it.

Signal always grows faster than noise (eg sky glow), so more/longer exposures improves signal, and required more in urban skies than dark sites

More subs and or longer exposures also allow more use of the deconvolution and sharpening filters, youd be amazed at how sharp you can make a 30hr (total exposure) fuzzy image before it gets noisy. Thats why the worlds best go nuts with megadata (apart from fancy rigs/dark sites). I dont go this far, but a good 4hrs on the Lum channel will get a sharp pic from subs with ugly eggy stars. Also, I process eggy stars to round (2 identical layers, in "darken" mode, shifted a few pixels between each other in the egg length direction) and minimise them by selecting/expand/feather and minimum filter.

My images are "overcooked" generally too, I still have much to learn.

In short, the worse the mount and seeing/glow is, the more data and time in CS you need to get a good image, its sort of cheating, but without more bucks on gear/moving to a dark site, its what you stuck with.

Overall, the main A07 experience was the need to plan well in advance. In Sky or Starrynight, only choose an object that has a suitable guide star with the FOV indicator, stuffing around jiggling the scope and endlessly rotating the cam randomly can suck hrs out of night and be very discauraging. I would now say an instrument rotator would almost be essential, and an automatic one driven by the Sky would be an absolute doddle, click and go.
The focuser/FR/A07/filter/cam and adaptors hanging off the back of the OTA can get unweildy too, its a major daylight job to set up, not something youd do everyday, with possible back focus problems (although I didnt have any on my LX200R OTA), and the need to rebalance.

Also, you need a new set of flats eveytime you rotate the cam, although I believe there is a way to get round that.

I forgot to mention before that the A07 also chases atmospherics (why it was designed in the 1 st place), it can halve FWHM, there is no other way given a rig that is tuned as far as it can be and seeing your stuck with.

With my DSLR, I never exposed longer than 5 min, although many do. I wouldve thought more than 10min would be not worth it in Urban skies, the glow just gets too bad up the histogram. The trick here is more subs, lots off them. 40, or even 100 makes all the difference. The signal is bigger and bigger compared to the skyglow, and even though the images may look similar after small or large stacks, much more processing can be done before noise shows up on big stacks.

Im no expert, Im just passing on what others have told me, and direct experience, I may have some of it wrong

Cheers
Fred

[jase (Jason)]

A mount with the lowest possible PE should be the starting point. AO is simply trying to counteract the affects of seeing to provide higher resolution images. It’s not meant to correct poor tracking, polar alignment or bad PE – while it will do this to an extent, it should not be depended upon.

Ultimately, what you want to be able to achieve is minimal guiding corrections. Everytime a correction is made, there is room for error. If you have a mount that is precisely polar aligned and tracks reasonably well, guiding corrections are kept to a minimum and you’ll be well on your way in the quest of pinpoint stars.

As Fred mentions, pempro can go greatly assist in PE correction. I’ve used pempro for about two years now and have redefined my PE at different stages. There is still room for improvement on the Losmandy Titan – a peak to peak PE of 2.34 arcsecs is ok, but could be considerably better. My intention is to get this down to around 1.7 arcsecs which I feel is still quite achievable. I may swap the DEC and RA worms to see if I get any further improvements.

You will note the first attachment shows a PE of around +/- 5.0 arcsecs, the second attachment shows the PE curve of 2.34 arcsecs. I measured the X/Y telescope axis error in MaximDL when guiding (at 4 second exposure times) which is shown in the third attachment. I will remeasure PE in pempro next clear night.

[Bassnut (Fred)]

Well said Jase, low PE should always be the starting point.

Your PE looks very low, well done. Whats left looks random and could well be partly due to seeing.