I am posting this info because my google searches turned up very little specifically about the issue I was having with my scope and the solution. This is on my Skywatcher Quattro 250P, but certainly applies to other scopes. 

I was experiencing astigmatism in the form of stars being "long" on one side of focus, and "tall" on the other side of focus. I am pretty certain that the issue was also causing some other issues with star shapes and slight doubling of only one of the two axes of diffraction spikes in some parts of my image. This was a recent development for me, but in retrospect, I was probably having a slight version of this issue for some time. Other symptoms were slightly oblong stars that I had attributed to guiding error being greater in RA or Dec in different situations, when really I was seeing slight focus error inside or outside of focus that was really showing slight astigmatism. I think that this stuff "fixed" itself during stacking since I tend to capture enough data that  the errors averaged themselves out. (Some of that is hypothetical and I have no concrete evidence that the latter observations/assumptions above are accurate...)

Almost all of the advice I could find online was that astigmatism was most likely caused by over tightening the mirror clips. I was pretty certain I had not over tightened my mirror clips and I was, in fact, correct.

What I had overtightened was the primary collimation stop-screws. This was slightly deforming the mirror cell and causing the mirror to not sit correctly in the cell.

These screws are slotted for a flat-head screwdriver. DO NOT USE A SCREWDRIVER, ONLY USE YOUR FINGERS TO TIGHTEN THEM!

I was, cleverly I thought, fine-tuning primary collimation by adjusting the adjusting screws to pretty-close, making the stop screws pretty snug, and then slightly tightening/loosening the stop screws to finalize the collimation which works for collimation but not in a desirable way, ultimately. I would estimate the torque that I had applied to the stop screws to be around 2.5-4 Nm, finger tight is 1-2 Nm (approximately). To be clear, I wasn't cranking down on the screws, but a little tight with a screwdriver is enough to deform the cell enough to cause astigmatism.

To be clear, collimation seemed perfect using the Catseye Collimation System, but I was still getting the adverse effects of astigmatism.

I also checked sensor tilt, focuser alignment, a different CC, confirmed backfocus to be correct on the CCs, all with no effect on the astigmatism result.

I ultimately made an artificial star with aluminum foil and a flashlight and put it as far from my scope as I could in my house. (~20 ft?) Once I had the scope set up with all of my extension tubes and an eyepiece, (this thing is not built for visual!), I was able to see the astigmatism clearly. I then backed the stop screws out until loose, and used only the adjuster screws to collimate the primary. I rechecked inside and outside of focus and the astigmatism was gone. I then snugged the stop screws a bit at a time with only my fingers, checking inside and outside of focus each time, until they were as tight as I could reasonably tighten them with only my fingers. The astigmatism did not return.

I remounted my scope on my mount and re-collimated with it pointed vertically, using only my fingers to tighten the stop screws. With the camera mounted last night, I observed no astigmatism, only donuts inside and outside of focus.

At this point I think I have the best stars I've ever had across my full FOV, even at f/3 with the Starizona Nexus .75x FR/CC. I think I had been a bit overzealous with tightening these screws since I got it, and that my mistaken collimation habit was successful enough for collimating with the Catseye Cheshire and autocollimator that I had only gotten more aggressive with it over time.

I did find this particular web page to be very helpful: https://www.loptics.com/articles/starshape/starshape.html

Hopefully this info helps someone else!


CS-

-Anthony
Edited ...
Like