X-axis shifts  

X-axis shifts

I recently tried to figure out what might be causing the x carriage to bind on the guides. With only the bottom bearing on a guide (rod), I tried to slide it left and right. It went left easily, but hung going to the right. After trying oil, smoothing with very fine sandpaper, and generally trying to learn if the problem had to do with a sharp edge of the bearing and very minor misalignment, I discovered some "flashing" of a hard rubber insert in the end of the bearing. That very small, thin sheet of rubber would fold over and jam in the bearing when the bearing was moved toward the flashing. Moving away from the flashing was not a problem.

I think this may not be uncommon. The flashing is so small it can be difficult to see. (I find black plastic is generally difficult to see detail in, so wish Prusa would use less of it. Maybe one of these days I'll print more parts of some different color, so that I can see the detail.) I trimmed the flashing off with a sharp blade, and also used a rat-tail file a little. Back on the rod, the bearing generally moved very easily, although it was somewhat sensitive to rotational alignment, with some orientations being smoother than others.

I'd like to say getting smooth travel out of the x carriage totally solved my shifting problems, but I'm afraid it did not. Next print attempt after a couple of 30mm cubes had both x and y shifts, big time. Among the recommendations I was directed to for curing such shifting was the statement that stealth mode can be associated with offset problems. Not sure I understand this, but I'll try it, soon as I get everything back together - I also reported in another topic that the nozzle disappeared.(!!!)

Both carriages move very smoothly. It's difficult to believe there is a problem with belt tension, although both are outside the recommended values, one high and the other low. I'd thought the belt might be induced to slip by high accelerations, and that stealth mode would reduce the likelihood that would happen. Maybe I'll have another try at reducing accelerations in Slic3r.