Astrophotography Thread

Nycturne

Elite Member
Joined
Nov 12, 2021
Posts
1,284
I’ve done a bit of astrophotography in the past, but bounced a bit off the hobby as it can be a bit of climb before you start getting good data when going from widefield with camera lenses to telescopes with long focal lengths. I also live in light polluted skies which means even more effort to get good results from your data, or more trips to do imaging which can be taxing.

I’ve also been on the waiting list for a rather high end telescope mount for a couple years now. My name got to the top of the list and I decided to go for it. I’ve spent the last week or so just learning the mount, and setting up a mini PC so the whole thing can be used as a self-contained astrophotography rig. Last night, the clouds cooperated and I got some time to do some actual testing. I wasn’t expecting to do much here, just ensure I can run through some of the tools, achieve focus with the cooled camera I have had sitting around for years, and see how well it tracks.

The thing that impressed me the most is how well this setup tracks. The mount is a Mach2GTO, which includes absolute encoders. This allows it to almost immediately catch and correct nearly all tracking error stemming from the mechanics of the mount. You can also build a sky model that helps account for other sources of error like polar misalignment (to a point), atmospheric refraction at different altitudes, and the like. I don’t quite have things setup for guiding yet, so I decided to try building a sky model and going unguided. When the short frames looked good, I decided to see how far I could push things with this 500mm focal length scope. The result is two 10min unguided exposures. No calibration, stacking or other processing, so there is still considerable noise. They are just stretched to show the detail available.

Also impressive was the pointing. The capture software I’m re-learning on has a feature to slew and center a target. It slews to where it thinks the target is, takes a short exposure and plate solves the image to find out where you are actually pointing, and uses that to adjust the mount a bit more and center the target. Most mounts will not always slew to a target perfectly, because of things like atmospheric refraction, or error in star alignment procedures. This thing always put the target dead center every slew. Never needed a follow up adjustment after the plate solve. I’ve used mounts that were considered quite good for the price, but none of them are in the same class as this thing. I’m still picking my jaw off the floor.

There’s certainly things I need to address. It took over 100mm of extensions on the camera to come to focus, which probably added some vignetting you can see in the Andromeda shot. The field curvature is pretty extreme in this specific telescope, so I need to look at a field flattener. It looks like there is some drift from polar misalignment. No attempt at fighting the light pollution was done. Focus is a bit off. But still, this is not a bad start.
 

Attachments

  • IMG_0357.jpeg
    IMG_0357.jpeg
    202.3 KB · Views: 36
  • IMG_0356.jpeg
    IMG_0356.jpeg
    286.5 KB · Views: 35
^ congrats! Looking forward to your developing images!

I have an old Meade Schmidt-Newtonian (10" F4 but a terrible mount!) and a camera I didn’t get round to using as we moved from relatively dark skies to a somewhat light-polluted neighborhood soon after. One day, when i get time…
 
I’ve done a bit of astrophotography in the past, but bounced a bit off the hobby as it can be a bit of climb before you start getting good data when going from widefield with camera lenses to telescopes with long focal lengths. I also live in light polluted skies which means even more effort to get good results from your data, or more trips to do imaging which can be taxing.

I’ve also been on the waiting list for a rather high end telescope mount for a couple years now. My name got to the top of the list and I decided to go for it. I’ve spent the last week or so just learning the mount, and setting up a mini PC so the whole thing can be used as a self-contained astrophotography rig. Last night, the clouds cooperated and I got some time to do some actual testing. I wasn’t expecting to do much here, just ensure I can run through some of the tools, achieve focus with the cooled camera I have had sitting around for years, and see how well it tracks.

The thing that impressed me the most is how well this setup tracks. The mount is a Mach2GTO, which includes absolute encoders. This allows it to almost immediately catch and correct nearly all tracking error stemming from the mechanics of the mount. You can also build a sky model that helps account for other sources of error like polar misalignment (to a point), atmospheric refraction at different altitudes, and the like. I don’t quite have things setup for guiding yet, so I decided to try building a sky model and going unguided. When the short frames looked good, I decided to see how far I could push things with this 500mm focal length scope. The result is two 10min unguided exposures. No calibration, stacking or other processing, so there is still considerable noise. They are just stretched to show the detail available.

Also impressive was the pointing. The capture software I’m re-learning on has a feature to slew and center a target. It slews to where it thinks the target is, takes a short exposure and plate solves the image to find out where you are actually pointing, and uses that to adjust the mount a bit more and center the target. Most mounts will not always slew to a target perfectly, because of things like atmospheric refraction, or error in star alignment procedures. This thing always put the target dead center every slew. Never needed a follow up adjustment after the plate solve. I’ve used mounts that were considered quite good for the price, but none of them are in the same class as this thing. I’m still picking my jaw off the floor.

There’s certainly things I need to address. It took over 100mm of extensions on the camera to come to focus, which probably added some vignetting you can see in the Andromeda shot. The field curvature is pretty extreme in this specific telescope, so I need to look at a field flattener. It looks like there is some drift from polar misalignment. No attempt at fighting the light pollution was done. Focus is a bit off. But still, this is not a bad start.

Long ago I was thinking about purchasing a decent telescope. And using MATLAB with their image processing and other toolboxes to enhance captured images. Life happened, and I live in an area where there's massive light pollution, so I kind of gave up on that.

Was wondering if you might be considering similar image processing to enhance what you capture. Matlab is cheap for a personal home license, as is their image and signal processing toolboxes. Their image processing toolbox is pretty deep and just $45 for home use.
 
^ congrats! Looking forward to your developing images!

I have an old Meade Schmidt-Newtonian (10" F4 but a terrible mount!) and a camera I didn’t get round to using as we moved from relatively dark skies to a somewhat light-polluted neighborhood soon after. One day, when i get time…

Yeah, I’m basically starting from scratch, just with a Ferrari instead of a Civic. We’re right next to Seattle so you can’t even see the Milky Way. But I am sorely tempted to try some narrowband. The mount can do it and auto guiding is going to be pretty clean on this mount.

Sounds like a fun scope though.

Long ago I was thinking about purchasing a decent telescope. And using MATLAB with their image processing and other toolboxes to enhance captured images. Life happened, and I live in an area where there's massive light pollution, so I kind of gave up on that.

Was wondering if you might be considering similar image processing to enhance what you capture. Matlab is cheap for a personal home license, as is their image and signal processing toolboxes. Their image processing toolbox is pretty deep and just $45 for home use.

Matlab is definitely not what I would expect to use for processing. There’s a few stacking tools that are free these days, and I’ve started playing with the stacking tools in Affinity Photo. I’m thinking of starting there for now as I already have a license and the stacking persona actually works pretty well.
 
Yeah, I’m basically starting from scratch, just with a Ferrari instead of a Civic. We’re right next to Seattle so you can’t even see the Milky Way. But I am sorely tempted to try some narrowband. The mount can do it and auto guiding is going to be pretty clean on this mount.

Sounds like a fun scope though.



Matlab is definitely not what I would expect to use for processing. There’s a few stacking tools that are free these days, and I’ve started playing with the stacking tools in Affinity Photo. I’m thinking of starting there for now as I already have a license and the stacking persona actually works pretty well.

I was thinking some of their image filtering and deblurring operations (just as one example) using blind deconvolution and other methods could be interesting. Long ago I used MATLAB and their specialized toolboxes for work-related signal and image processing tasks.

Just an idea, thinking they could be useful in amateur astronomy where imperfect captures could be made better...
 
Oh, processing this stuff is definitely a whole thing with how little SNR you have. It’s just a lot of the tools that do stacking have presets and tools aimed at DSO work. But it’s also a lot easier when you have a bit more than 10 minutes of integration to start with.

Planetary really does benefit from deconvolution in particular though. It’s a whole different game than DSOs.
 
Some playing with Affinity Photo as a stacking and processing tool. Surprisingly good results from the stacking persona. I didn’t get to take flats for this so I had to deal with the gradients and sky glow manually. Not too bad from polluted skies, but not sure I like the color balance. Should probably be a little more red. Still, for about 2 hours of “junk data” I think it isn’t too bad and demonstrates Affinity Photo is certainly usable for now.

Fingers crossed I get time with the new scope this weekend to start putting the new process all together and get some real time on a subject.

North America Take 3.jpeg
 
Some more testing across a couple different telescopes. I’m playing with a Trial of PixInsight now too. It’s a bit dense, but the results are much easier to work with. It’s kinda impressive how much difference a couple of sessions makes…

The rig itself is getting more and more automated. Still need to pick up a few things like a dew heater, and do some vignetting tests with my Sony A7 camera. I want to go mono at some point so I can do some narrowband work, but haven’t decided if I want to go 35mm or APS-C on the mono camera. Depends how well the telescopes I have can illuminate the 35mm sensors available. PixInsight has some tools that can help me measure what’s going on pretty quickly here, without wasting a clear night doing it.

Still only a couple hours per target, and I’m still honing in on a good ”beginner” flow for processing, but it’s looking pretty nice. I’m pulling together something worth at least looking at and learning from much faster than I have in the past. The background removal on the North America Nebula in my latest session is a little aggressive, but the contrast may well be worth it with this little data.

IMG_0361.jpeg

Master Integration ABE (Proc2).jpeg
 
Back
Top