GEM Image Acquisition Physics Question

[jkmorse]

Hey Folks,

Looking for some physics help (though good old fashioned actual experience in the field is welcome too).  As noted below, my primary image acquisition rig has a Planewave CDK mounted on a Paramount MX GEM.  I have the ability to mount a second unit above the CDK by attaching my STF8300 with a Canon lens to a bracket on the CDK.  This way I should be able to add wide field captures to the narrower image through the CDK without needing to run a second mount.

Here's the question.  Assuming autoguiding along the image plane of the CDK (using a MOAG attached to my F16M, which in turn is attached to the CDK), is there a practical limit to how far off-axis I can point the 8300 and still have it track properly.  I can solve this by trying it in the field, of course, which I will do the next time clear skies meet up with no moon, but I would like to see if I can get an answer in advance if possible.  As I have mentioned before, I love to dabble in physics, but as a History major in college, it is hardly a specialty and I can't figure this one out on paper, though by drawing diagrams it seems like I should be able to image well off-axis and still have a good image.

Thanks in advance for any light you can shed.

Best,

Jim

[Geoff]

In a perfect world(no atmospheric refraction, exact polar alignment, no periodic error etc, etc) the misalignment could be anything up to +-180 degrees. In practice, with a paramount and reasonable polar alignment and a camera lens focal length I think you could misalign as much as you want and not see any tracking error.
Geoff

[jkmorse]

Geoff,

Thanks!! though I am seriously doubting my ability to get any where near perfection.

Best,

Jim

[troypiggo]

I'm not lucky enough to own or use such a high-end mount, so not sure how much of my experience is relevant, but...  I have a Tak EM-200, and recently mounted a ballhead and 5DII/135L on top of my refractor.  I wasn't even shooting that far off the axis of where my refractor was pointing.  But when I rotated the camera for composition using the ballhead, it definitely throws the balance of the scope/mount out.  Enough to necessitate having to recalibrate guiding every time, and even then the guiding adjustments were working much harder than when everything perfectly balanced.  I didn't notice it so much in the DSLR/lens images as the focal length was so short, so errors in guiding are quite forgiving at that image scale.  But if I was imaging through the refractor at the same time, I am pretty sure I would have noticed it and gotten lower quality images due to less-than-perfect guiding than if didn't have the camera on top.

If I could some how rig it up so that as the DSLR on top didn't affect balance, I'm sure it would be fine.  As I said, at that image scale it's very forgiving.  I think the arcsec/pixel I was imaging at was more than the guidescope/guidecam!

The balance may not be as much of an issue for you, considering the much higher capacity etc.  Only trial and error will show.

[jkmorse]

Troy,

Thanks, but now I am worried about another problem.  At the image scales we are talking about, drizzle becomes really important, but then you need to dither your images and unless I physically move the add-on setup (which is likely to affect the base OTA), then I can't think of a way to dither on a piggyback. 

So likely not going to work as I hoped.

Best,

Jim

[pfile]

if you are guiding, you should be able to dither…

not sure why it's any different with a piggyback… you may need to adjust the absolute size of the dither in arc seconds (has to be pretty large for a wide field image) but other than that, why won't it work?

rob

[jkmorse]

Rob,

Oh, I am dithering, but doing it with my main scope, which is taking 10 minute subframes.  With the piggyback, I will be using a fast f2 Canon lens and that only needs 2 or 3 minute iterations.  And I have no idea how to make the piggyback dither separately from the main scope.  Personally, I blame Newton.

Best,

Jim

[troypiggo]

I think it comes down to your image scales.  What are the arcsecs/pixel of your guider, main imager, and the piggy back?  What lens are you proposing to use on the piggyback?  You may find that if you reduce your dithering amount to the bare min needed for the main imager (CDK), the piggyback lens image scale won't even notice the dithering.

[pfile]

ok that makes sense. i guess if the image scales are too different then a dither for the wide field might take the CDK right off the target…

i suppose you can just use them separately and then crank the dithering when the wide field imager is going.

rob

[jkmorse]

Yeah, I was caught this weekend by the reverse.  I usually use an SBIG Sti on a MOAG for guiding but decided to test using the Sti guider system which uses a little 100mm lens.  But that meant that when I was guiding at a relatively sloppy 0.5 pixels, my CDK, at 2541 focal length, was moving around up to 12 pixels!! and even for guiding at 0.1 pixels RMS, it was still moving 3 to 4 pixels.

As usual, seemed obvious after the fact but really caught me by surprise until I ran the numbers to see just why my stars were so smeared.  Glad I did it with a full moon so nothing worth saving anyway.

Jim

[mschuster]

Jim,

Here is a calculator that might help.

http://celestialwonders.com/tools/polarMaxErrorCalc.html

Field rotation scales roughly linearly with exposure time and off-axis angle. Double the exposure time or double the off-axis angle, then your allowable polar alignment tolerance is half as large. Target declination is a factor also.

On my setup, with 40 minute subs, off-axis angle is small (about 3 degrees from OAG guide star to farthest frame corner). Calculator say 3.5 arcmin is allowed error for a 60 degree declination target keeping field rotation under 10 microns. This is probably in the ballpark. In my tests, a 2 arcmin PA error gives excellent results, but a 10 arcmin PA error definitely shows field rotation.

Mike

[jkmorse]

Mike,

Thanks for that!! I will run some tests and plug into the calculator next time I am in the field.

Best,

Jim

[jkmorse]

Mike,

One other question.  You say that you use 40 minute subs.  Doesn't that result in over exposure of stars and background.  Either that or, at least for the background you have amazing skies to work with.

Jim

[mschuster]

Doesn't that result in over exposure of stars and background.

Bright stars yes, background no: 3nm Ha filter.

Mike