John Hayes:
Daniel Erickson:
Thanks, John. Your explanation is really clear and the charts help. For the moment, let's assume that the software is fine (I'm using a pre-made script in Siril) and that the issue is optical... Being new to all this, how do I go about checking for and fixing optical issues? I don't expect a long treatise of course--you've been generous enough--but if you can point me towards something to read/research, I'd be grateful. Thanks again.
There is a lot to identifying and fixing optical problems but broadly speaking, the problems fall into two classes: 1) Aberrations due to fabrication errors or design limitations and 2) Alignment induced errors. Problems in the first category can be very difficult to resolve without getting the manufacturer involved. The trick there is to know when you've correctly identified a problem that really is due to a manufacturing error. Simply because you might have a scope with a sophisticated design that was very expensive does not mean that it will produce perfect images. Understanding how to test and evaluate optical performance is a big subject but one good place to start might be with Suiter's book on star testing.
Issues in the second category include mostly misaligned, tilted, or improperly spaced components, although it could include something like a multi-element objective with lens components that are misaligned with respect to each other. These are things that are ultimately fixable without having to remanufacture parts. In this case, you have to go through all of the specs and measure everything with respect to the specified spacings. Suiter's book can also help determine what aberrations you might have though star testing. This is an area where hooking up with someone who has more experience with this stuff can be really helpful. There is almost no way that I can give a short answer here that would be of much help without addressing a very specific problem.
Finally, I've asked a couple of times about your registration accuracy but forgive me, I haven't seen a clear, definitive answer. Blinking through your stack, zooming into the center and four corners while you cycle through the aligned stack can be very revealing. Have you very carefully examined each of the aligned subs before you did the image integration? This is where small errors that are easy to miss can turn into very big problems. Believing that the stacking process is correct is not the same as actually carefully examining the data. Perhaps I'm misinterpreting your answers (and forgive me if that's the case,) but this is a very important step when you run into problems like this.
Best regards,
John
Thanks, John. I appreciate the time you put into this. I was specifically wanting to know about the optical possibilities because in a PM I began working with someone else who knows Siril quite well about the registration and stacking issues. I wasn't trying to dodge your question.
But, since you mention it, please remember that I am an ignorant beginner and--quite honestly--I'm feeling rather beginner-stupid right now. As such, I tend to blunder my way through to an image. I feel quite confused and I'm also frustrated, because I want to do good work and I can't learn fast enough to fix problems, it seems... Still, my process, so you understand it, is to shoot my subs (after focusing, etc. etc.). Once I have those files I go through each one looking for bad ones--ones with airplanes and satellites, odd lighting or just plain strange looking frames. At my level of expertise I suspect I don't notice much more than the obvious, I admit. I get rid of every one that doesn't look perfect to me at this time in my own understanding of what constitutes 'good enough'. I then put all my calibration frames (darks, flats, flat darks) in folders with the lights in theirs, and execute a Siril script that *automatically* pre-processes the files, registers and stacks them. Siril is nothing more than a black box to me at that point, so if something is wrong there, I wouldn't know it--hence working with a Siril expert on the matter. The resulting .fit file I then do the photometric color calibration and background extraction on... and here we are.
One reason I thought it might be an optical issue is the star you labelled "registration error". Those occur from the edge of the frame to the center--i.e., everywhere. Perhaps the plate on the back of the scope, which has four tiny set screws, is misaligned...but I'm clueless about how to diagnose that and to know the best way to fix it (turn the screws, that's it). That particular star also has a pink core. You would say then, that it, too, is also overexposed? As for the stars you label as 'coma', I thought my RedCat (quad apo petzval) might actually have avoided that issue entirely. I'm wondering whether it's possible that the 'coma stars' are actually mis-registered (and overexposed) so that one frequency of light is brighter and gives the appearance of coma... just a thought.
Perhaps in the interest of my sanity and your time, I will shelve this data for the time-being. I need to shoot some more data on the same target and same general set of conditions to see if whatever this is recurs. I also want to try decreasing exposure. I've had some ongoing discussions with
@Björn Arnold . I may learn something from my exploration of Siril, too. BTW, you mention "Suiter's book". I don't know what reference you're referring to.
Thanks for all your help John. I don't know that I'm any closer to knowing what's going on at the moment, but I am learning a lot.
Kindly,
Daniel
