Like many of us, I'm trying to find a silver lining to the "shelter in place" orders still in effect in the San Francisco Bay area. One of the benefits is that I now have lots more time for my imaging. Even that, though, is not an unmitigated good. I can't get to my observatory in northern Napa county. The last time there were fires up there, I actually brought the equipment home as a precaution, so everything has just been sitting in my garage for a while. I decided to find out just what is possible from my Bortle 8/9 skies here in Oakland, CA. I also wanted to determine how to optimize the configuration for my mount, camera, telescope, and processing. For those of you who don't already know it, getting a good deep sky image from any given location is about incremental improvements. Improve the accuracy of your guiding. Improve your collimation. Improve your pointing accuracy. Address the cooling issues in your optical tube. Improve your polar alignment. Improve your focusing accuracy. Determine the offsets for different filters (even parfocal filters). Improve your calibration files (more/better). Any one of these little improvements has almost no effect at all on your results, but if you keep plugging away making little changes, you will see your FWHM numbers (resolution) getting better and better until you are truly seeing limited. You'll see your signal-to-noise ratio improving allowing you to go deeper in a given amount of exposure time. Add them all up, and things start to look pretty good, even from where I live.
So, what have I done? First, I installed a permanent pier in my rose garden. That way I can leave things setup over night (covered as required for bad weather) without having to redo polar alignment each time I go out. Then I spent time getting a good enough polar alignment that ten minute unguided subexposures show no visible eccentricity in the stars. Then I built out my horizons so the mount would know where it could/couldn't slew. I'm in the middle of the city of Oakland, just a mile or two from downtown, so not only is skyglow a huge, huge problem, but I also have very limited horizons due to neighbors. Due South I can get all the way down to 20 degrees or so, but to the North, East, and West my local horizon can be as high as 45* or so. That doesn't leave much sky available, so imaging time is limited on any given night for any given subject, and I am very limited in what I can choose to capture. Such is life.
After building out the local horizon, I taught the mount how far past meridian it could be without the telescope hitting the pier. The utilities from Astro-Physics let you do this for every different part of the sky. This dramatically reduces the need for meridian flips, allowing me to go at least two hours past the meridian on must subjects. Pretty cool. Next, I built out a high precision pointing model so the scope can track more accurately as subjects move across the sky. It accounts for tube flex, atmospheric refraction, errors in polar alignment, etc. That's the part that allows ten plus minute unguided exposures even at moderate focal lengths. Finally, I measured and compensated for backlash in my focuser and built out a good set of V curves so I could automate focus. I don't have a thermometer hooked up to the telescope, so no way to do temperature compensation at this point, but with the automation, it's easy enough to just re-focus every hour or so. Finally (at least for data collection), I worked out a routine that gets decent quality twilight flats. I don't have any easy way to do daytime flats, so I'm stuck with T shirts at dusk, but I've now got a routine that seems to work consistently. Do it for a couple nights, and as long as I don't rotate the camera I'm good to go. Periodic error? Non issue since the mount has absolute encoders. No guiding needed. Pretty cool. Collimation? It's about as good as one can get. CCD Inspector reports no tilt and just a bit of residual field curvature. That's about all I can hope for in a fast f/3.8 system. The scope even has a thermal blanket to prevent heat plumes and differential cooling.
So, I should be good to go, yes? Well, everything is about as good as I can get it, but I'm still in Oakland and I'm still trying to image looking out over people's hot rooftops. So getting anything better than about 2" FWHM is highly unlikely, and going really deep is virtually impossible. But it's still a lot of fun, and it's really nice feeling like the equipment is working to its full potential for the first time.
Here is my first complete image since setting up here in Oakland, CA. M101, obviously. This is one hour each of R, G, and B data merged with three hours of luminance. Not nearly as deep as I would like and definitely lower resolution than I had hoped for, but you take what the skies allow. I may add more data to this and see whether I can improve SNR a touch, maybe draw a bit more out of the shadows. I ran through some of Adam Block's tutorials in his "fundamentals" series on Pixinsight to get to this result. I'm actually pretty happy--probably the best technical image I have produced in a couple or three years, and the best I have ever taken from here at home.
All processing was done in PixInsight. Subs were captured in Maxim DL. APCC was used to run the mount. FocusMax took care of keeping things sharp. The scope was a 305mm Riccardi Honders. The mount was an AP1200GTO AE. The camera was an STXL-16200 from SBIG. Any weaknesses in the image are my own, not the equipment. Overall, pretty happy with this!