So this was my first print (and printer) attempt and the results could have been better... Any suggestions from those with experience as to whats causing/how to fix the issues? I am using the 2s dual setup, but only hotend 1 was used and the fan shrouds are currently not on. From what I have read the shifting is typically caused by a mechanical issue, but it didn't really show up until the later portion of the print (about an hour into the print) and a little at the bottom and the splitting is caused by incorrect temps or layer thickness, but I was using the suggested default... Any suggestions would be much appreciated. Thanks.
Could be a corrupt gcode file, try using a different SD card writer/reader.
Could also be too much current going to through the stepper driver the direction the layer shifting is happening in causing it to over heat and shut down temporarily. Try reducing current by turning the trimpot. Turning the trimpot counter-clockwise will lower the current. *Turn slightly and check (1/8th turns)*
Did you manage to fix your issue. It seems I'm now running into this issue to. Turning the trimpot on the Y-Stepper motor hasn't had any effect for me. Although, I'm getting the opposite result: My first few layers are shifted but it does pretty good at not shifting most of the rest of the layers.
Yes and no... I *slightly* adjusted the y-stepper trimpot and took a little tension out of the y-axis belt (1-2 teeth worth) as I think I may have over-tightened it when building. The next Benchy I printed had room for improvement but was, in my amateur opinion, "acceptable". Since then I've had a number of other issues though: non-working DSE servo, seemingly random z-leveling, random hotend temps (set to 195, displays 195, but heats to the point of burning the filament), randomly shifting 4 inches along the x-axis while printing, "printing" while just repeatedly bumping the x/y endstops, small shifts/jumps on the x-axis during prints, and prints that just randomly stop. I've used the same compiled g-code for all the print attempts so the file shouldn't be the issue. So far I have had the one "acceptable" benchy out of 6 or so finished attempts.
If you have a multimeter you will be able to fine tune the TMC2100 drivers:
You can measure the Vref by placing the positive probe on the trimpot and the negative probe to the negative terminal from the power supply on the main board.
For the X axis (platform) you can set it between 0.95-1.05v and for the Y-axis (hotend) you can get away with less, between 0.6-0.7v. If an axis is still struggling, check you don't have any resistance on the axis when powered off and moving it by hand.