As I continue to build on this stack, I've started adding star separation to the Siril processing workflow. I may have overdone it a bit this time 😄. Working towards a 30 hour stack.
I've learned so much from this subreddit, so I wanted to share something I learned:
Since I got my S50, I've been using an Innorel F20 fluid head as an EQ mount (it's the same thing as the Seestar TH10) and I've found it to be very fiddly trying to dial in the perfect EQ angle. I could never do more than 20sec exposures without dropping a large percentage of frames (>50% for 60sec), and with star trails in the frames it keeps. Today I'm trying out an iOptron 3327 EQ wedge that I got from milehighastro.com for about the same price, and was able to pretty easily get EQ alignment dialed in to 0 degrees, and suddenly I can do 60sec frames! I'm backing it down to 30 sec for my plan tonight because I don't want to push my luck, but it's an instant huge upgrade over the Innorel head.
If you're struggling with long exposures and using a TH10 or similar, my advice is just get yourself a wedge and be done with it. I chose the iOptron over the Sky-Watcher because it was a bit less expensive, and the knob to secure the Vixen plate is lower profile. That said, the iOptron did not include a plate and I had a hard time finding one for a reasonable price that would work with the S50. I ended up getting one from Amazon, "YUEOCT" brand, that has an Arca Swiss clamp on it, and just left the Arca Swiss plate on my S50. Works great.
Here's three images of M13 processed from the same .fit files. 1:04 integration time using 20s exposures (193 total frames).
I then processed all the stacked images using the same workflow in Siril:
These captures aren't great, as the stars are not pin sharp, especially at the edges. I need to get better at polar alignment. I also still have a lot to learn about processing images in Siril.
Anyway, I think the best result comes from the in-device stacking, which is too bad because that is a pain in the ass, and very slow. The ASIDeepStack did the worst, and provided the smallest usable image area (these are all cropped as loosely as I could manage).
I suspect I could improve my results with the Naztronomy script. Anyone got any tips, or links to a good tutorial? What do you all think?
