I had a similar experience as yours with my 1100 mount shortly after I got it. I am also strictly a mobile user, setting up and taking down the mount in my back yard every time I use it. I am using the old 8-inch pier that came with my A-P 800 mount that I purchased in 1992, and I noticed the polar axis would sometimes shift during an observing session.
What I found in my case was that the top of the 8-inch pier was not completely flat. When assembling the mount I noticed a distinct wobble when I placed the base of the 1100 mount onto the top of the pier. I found that the top of the pier tube was not flat, and that the machined base of the 1100 mount was only making contact with the top of the pier in two places. I estimated that the base of the mount was tilting as much as 0.035". I could tighten the bolts holding the base of the mount onto the pier, but this was not enough to keep the assembled mount with the scope and counterweights from shifting under a load.
The solution was pretty simple. I coated the top of the pier with layout dye, placed the base of the mount on the pier, rotated it back and forth slightly, then removed the mount base. The high spots showed up as spots on the top of the pier where the dye had rubbed off. Using a file, I carefully filed the high spots (light and gentle touch on aluminum !!), recoated the top of the pier with layout dye, and repeated the process. It took about 45 minutes to get the base of the mount making uniform contact completely around the top of the pier. After doing this the episodes of shifting polar alignment stopped.
I have long suspected that another potential source of polar shifting is due to the generous clearance between the bolt holes on the top of the pier and the 5/16-18 bolts that are used to attach the base of the mount to the pier. As long as the load is centered over the base of the mount, the assembled mount with scope and counterweights is is not likely to shift, but I think it would be better to have some alignment pins that positively lock the top of the pier and mount base in position. The current configuration of the 1100 mount uses six bolts to attach the base of the mount to the pier. I would like to see three of the bolts replaced with tapered dowel pins. This would provide a rock solid connection of the mount base to the pier that will not shift (assuming the top of the pier is flat). The only problem with tapered dowel pins is that they can really get stuck. A method of removing them has to be devised, but I'll bet that Roland's outstanding machinist can devise a very clever way to do this.
I hope that this helps. Good luck in solving your problem.