A few weeks ago, I posted a thread where I noticed that my Pegasus Astro Ultimate Powerbox V2 was reporting temperature incorrectly, which potentially affected the tracking model. I also noticed that I had refraction correction disabled. Correcting the temperature and enabling tracking correction improved my situation, but still wasn't giving me completely round stars.
We've had mostly overcast skies since then, but we did get one clear night a few nights ago. I tried an experiment where instead of creating an all-sky model with APPM, I made a model with points only along the declination for my target object. Normally, my all-sky model isn't that dense, with between 150 and 300 points total. To do the declination only model, I did samples every two degrees of RA. I limited hour angles to only those where my target would be up during darkness hours. Since my object transits fairly early (and it took me some time to set things up the way I wanted), I ended up with only 4 points on the eastern side, with 44 points on the western side.
When I made my first 10 minute exposure (at 0.88 arc seconds per pixel), the RA tracking was off. Actually, that's an under statement. The RA tracking was so poor that I wasn't getting elongated stars. I was getting short star trails. My assumption is that with only 4 points on the east side, there was significant error in the model. When I looked at the model details, though, I saw that the Index Hour Angle for the east side was *really* strange (think minus 17,000). As a test, I disabled the Index Hour Angle term from the model.
When I did that, the stars were *really* round for my 10 minute exposures. They were easily as good as I would have seen with tight guiding. The first exposure after the meridian flip was similarly excellent. Since the west side value for the Index Hour Angle looked pretty good (around 1), I re-enabled that term. After doing that, I saw some star elongation in the RA direction, so I disabled it again and got perfect stars.
So until APCC Pro support declination arc tracking, I think that I'll keep exploring tracking models custom created for specific targets. My main goal is to get the kinds of results that I was seeing and with high reliability. I'm encouraged by what I saw the other night.
For reference, the mount is an AP1600 with Absolute Encoders. The scope and camera is an AP130GTX with a QSI690. I know that I can easily automate this setup with OAG guiding enabled. But I want to explore what's possible unguided, since it's a bit more efficient from a time standpoint, since it doesn't need to reacquire the guide star after a dither. Also, I have a new ASI2600MC Pro that I want to try. If I can get comfortable with unguided imaging, I'll set it up with no OAG or guide scope.