DSO long exposures without motorized mount?

[Mr_Timn]

I'm working on a school capstone project and decided I would try branching into astrophotography as observational astronomy has always interested me, but not so much astrophotography. Since I'm mostly concerned with observational astronomy, I just have an 8" alt-az manuel dobsonian, and I got an orion starshoot 2 camera just to try things out and see what I could image. I was wondering if it would be possible to get any detail out of a DSO if I were to take exposures in increments and move my telescope to center the object between exposures. Any otherimaging  methods or object suggestions that I could image for my school project would be greatly appreciated. Thanks!

[Shibby]

I have to say, it'd be very difficult - drift will start showing in a matter of seconds.

Have you considered lunar and planetary imaging? There's plenty of fun to be had there with a dobsonian. It's also quite well suited to your camera and its high frame rate. There are plenty of guides on this forum and across the web on how to get started with lunar/planetary.

[celestron8g8]

I have a C8 on a CG5 with tracking motors . When I bought it new and before I added the motors I did several nights observing . When I did not move the knobs drift was fast enough that objects went from one edge to the other in a matter of a few seconds using my 25mm without a barlow . My point is drift will be fast enough that star trailing will be immediately . Planet shots will be blurred and DSOs' such as galaxies and nebs would be a blob . With a dob and camera mounted not sure the weight would even keep it steady for a 2 sec shot . 

[Mr_Timn]

15 minutes ago, Shibby said:

I have to say, it'd be very difficult - drift will start showing in a matter of seconds.

Have you considered lunar and planetary imaging? There's plenty of fun to be had there with a dobsonian. It's also quite well suited to your camera and its high frame rate. There are plenty of guides on this forum and across the web on how to get started with lunar/planetary.

I plan on imaging some planetary objects, but I was wondering if I were to align each star in persay, the pleiades, each fram would stack nicely and look reasonable, although the edges would probably look a bit odd. Even if I could increase some detail with 4 or 5 frames, I'd like to just to get the experience. Thanks!

[Shibby]

The problem is, your frames (sub-exposures) would have to be so short that you wouldn't pick up any detail in the faint DSOs. Depending on your focal length, they might need to be a fraction of a second to avoid any trailing within the exposure. Perhaps you could still pick up some of the bright stars / clusters, so you could always give it a go.

Planetary images are stacked from videos taken at (for example) 30fps so shouldn't be subject to any noticeable trailing.

[Mr_Timn]

10 minutes ago, Shibby said:

The problem is, your frames (sub-exposures) would have to be so short that you wouldn't pick up any detail in the faint DSOs. Depending on your focal length, they might need to be a fraction of a second to avoid any trailing within the exposure. Perhaps you could still pick up some of the bright stars / clusters, so you could always give it a go.

Planetary images are stacked from videos taken at (for example) 30fps so shouldn't be subject to any noticeable trailing.

With that in mind, I'm curious as to what the difference is between stacking jupiter's moons, and a bright star cluster. They both seem pretty similar. I'll just have to try it out myself

[Mr_Timn]

Assuming imaging DSO's with my current setup wouldn't be worthwhile, are there any other methods or equipment that I could use to get some kind of increase in detail? I'm open for any suggestions.

[Mr_Timn]

42 minutes ago, celestron8g8 said:

I have a C8 on a CG5 with tracking motors . When I bought it new and before I added the motors I did several nights observing . When I did not move the knobs drift was fast enough that objects went from one edge to the other in a matter of a few seconds using my 25mm without a barlow . My point is drift will be fast enough that star trailing will be immediately . Planet shots will be blurred and DSOs' such as galaxies and nebs would be a blob . With a dob and camera mounted not sure the weight would even keep it steady for a 2 sec shot . 

Correct me if I'm wrong, but if the stacking process of planetary imaging already aligns each frame to compensate for drift, then wouldn't the same aply for a DSO, even if it's a short exposure. I don't plan on getting amazing hubble quality images, just a little extra grey poof. I'm pretty sure there's a difference between DSO and planetary imaging, but I'm having a tough time wrapping my head around it if the rate of drift and exposure length are the same for both in this case.

[Shibby]

The difference is that lunar/planetary imaging only requires very short exposures (e.g. 30 fps) whereas DSOs are generally much, much fainter so longer exposures are required.

As I said, you could try the brighter clusters and you may be able to pick up the brightest stars in them, but even these are fainter than the planets and the Moon.

[Mr_Timn]

3 minutes ago, Shibby said:

The difference is that lunar/planetary imaging only requires very short exposures (e.g. 30 fps) whereas DSOs are generally much, much fainter so longer exposures are required.

As I said, you could try the brighter clusters and you may be able to pick up the brightest stars in them, but even these are fainter than the planets and the Moon.

Like I said, I'm relatively new with this, so does that mean one frame of video of a brighter star cluster would be fainter than observing it through the eyepiece? If that's the case, this makes a lot more sense. Thanks for all this help!

[Shibby]

10 minutes ago, Mr_Timn said:

Assuming imaging DSO's with my current setup wouldn't be worthwhile, are there any other methods or equipment that I could use to get some kind of increase in detail? I'm open for any suggestions.

While I believe it is possible to get a motorised alt/az mount for a dobsonian, the fact remains that they're really designed for visual astronomy so I'd question whether it's worth spending money down that road. For DSOs, the equatorial platform is really the only way to go.

[Thalestris24]

Hi

You could ditch the Dob and do some widefield AP with a camera and wideangle lens - see our widefield imaging section. See also 

Having a dark sky is a great help.

Louise

[happy-kat]

You beat me to sharing Forest's video :-)

[vlaiv]

Actually there is a way of doing it but results will certainly be underwhelming compared to tracked AP.

It also involves quite a bit of math and maybe some fancy programming. It can be done with available free software but it will be slow as each sub needs to be processed in particular and individual way. Process can be automated with programming but I'm not aware of any software package that does it out of the box.

Idea would be as follows:

1. Center object (probably via finder), do exposure of certain length. Exposure length needs to be such that stars don't move on sensor more than given distance - this distance will impact width of final image, if stars move 1/4 of sensor - usable final image will be 3/4 of sensor width.

2. Go to step 1 for given number of times.

3. Apply regular calibration steps

4. For each frame, create kernel that consists of single line in direction of star motion on sensor and of that exact length. Kernel should have uniform distribution of values / brightness. Deconvolve each sub with its kernel. It is probably best to use RL deconvolution.

5. Stack deconvolved subs normally.

[Mr_Timn]

thank you everyone for all the help and going above and beyond in trying to help me, but all I want is to get something that would just be similar to what you'd normally see through the eyepiece, just to use as an example for a class project, nothing fancy. I think astrophotography is great, but I'm not all that much interested in taking amazing long exposures, nor do I plan to in the near future, I just wanted to know if I could get a tid bit extra detail instead of a single image or frame of a brighter dso. Thank you all for the help!

[alan potts]

I feel the best or at least one of the best shots of Saturn I have ever seen was taken by one of our friends down-under. He shot Saturn with his camera phone holding it at the eyepiece I believe, the downside for everyone, his Dob is the size of the average coffee table at 28 inches.

Alan

[Stub Mandrel]

8 hours ago, vlaiv said:

Actually there is a way of doing it but results will certainly be underwhelming compared to tracked AP.

It also involves quite a bit of math and maybe some fancy programming. It can be done with available free software but it will be slow as each sub needs to be processed in particular and individual way. Process can be automated with programming but I'm not aware of any software package that does it out of the box.

Idea would be as follows:

1. Center object (probably via finder), do exposure of certain length. Exposure length needs to be such that stars don't move on sensor more than given distance - this distance will impact width of final image, if stars move 1/4 of sensor - usable final image will be 3/4 of sensor width.

2. Go to step 1 for given number of times.

3. Apply regular calibration steps

4. For each frame, create kernel that consists of single line in direction of star motion on sensor and of that exact length. Kernel should have uniform distribution of values / brightness. Deconvolve each sub with its kernel. It is probably best to use RL deconvolution.

5. Stack deconvolved subs normally.

Astra Image allows you to apply 'motion blur' deconvolution by creating such an elongated kernel.

The results are far from perfect, but it is a useful tool when you have subs with a consistent drift (such as that caused by selecting lunar tracking rate, not that i would do such a thing  ).