To get that much movement in your pics you should be able to see the stars drifting in Phd. Find a bright star, stick your mouse pointer on top of it, have monitoring turned on but don't click on the PHD tracking button. If its moved more than a few mm on the screen within a few minutes you're going to be pushing it to get good pics.
From some of those I think you'd find the drift was >1 cm every few mins.
I still find my polar alignment can be bollocks when I set up in an unfamiliar remote site, even doing tricky webcam routines. Probably 'cause I get lazy at home having marked the pavers for the mount
