Last night was clear enough to do some quick imaging. Now that it's galaxy season, I changed the scope form my trusty 160EDF to the longer focal length 10" Mak-Cass. I used our CCDT67 compressor to bring the focal length down to 2500mm. The image posted below is a quick snapshot of 3 x 1200 sec with the QSI camera. The main purpose last night was not to take pretty pictures, rather it was doing ongoing testing of the Mach2 mount and the modeling software.
Two problems occur when using long focal lengths. First, is how to guide properly at 2500mm in order to capture fine galaxy detail. The best way is to use off-axis guiders. That brings up the second problem, which is the small field of these guiders. Sometimes there are not any bright stars to guide on. Therefore long guide exposures are needed, but then the mount must track very accurately between exposures. That means polar alignment must be spot on, otherwise you get drift between guide exposures. If you are setting up in the field, you don't want to spend a lot of time drift aligning to get perfect polar alignment.
I wanted to see what kind of results can be achieved with poor polar alignment. Last night I offset the azimuth axis so that I was getting about 2 arc seconds per minute drift in Dec. (RA was not drifting much). The guide star that was available needed at least 5 second exposure in order to be usable. With that much Dec drift, the guider was struggling and I achieved only 0.35 arc sec RMS tracking in Dec, while the RA was producing around 0.12 arcsec.
The solution was simple: I selected the Drift Model in the keypad, measured about 10 minutes of drift data with the guider keeping the guide star on the crosshair. Then simply turned on the Drift Compensation. The guider settled down almost instantly, I was able to dial the aggressiveness down to 60%, and the guider graph started to show Dec tracking of 0.12 arc sec RMS, same as the RA. There were periods of calm where there were no corrections being sent at all for 1 - 2 minutes at a time.
This process is fast and essentially foolproof and can be repeated every hour or so to build up a drift model for that object that can be used on subsequent nights. It can be done on the fly while you are imaging without disrupting your imaging session.
Here is the result from my heavily light polluted observing site at the AP factory (remember, it's just a quickie 1 hour snapshot):