Last night was the first clear night in a very long time, and although it wasn't very transparent, I did get to work out some of the kinks in the system.
I ran PemPro a bit to measure the backlash, and it was signifcant in DEC.. upwards of 40arcsec. Long story short, I took the DEC motor housing apart and discovered that the spur gear set screw wasn't seated properly, and the spur gear was shifting upon motor reversal, introducing the backlast. I fixed that up, put it all back together and the DEC backlast is greatly reduced.. not perfect, but a LOT better than before.
I also played a bit with PHD guiding, and it works a LOT better than MaxIm guiding. It means going back to much more user intervention during imaging sessions, but I'd much rather that than continue to be frusterated with MaxIm. My stars still are not perfectly round so I'm convinced that there is something wrong in the imaging train, but for the meantime, I'll be using PHD and taking images manually.
The only real image I took was an hour worth of M33. I thought I was shooting HA (as it's a 90% moon), but forgot to plug in my filter wheel, so I was really shooting Lum.