MK4 X Layer Shift
I had this crazy X layer shift on a long overnight print on my kit-built MK4. The belts and pulley setscrews are tight, manually moving the X axis is smooth, the magnetic sheet didn't move, and adhesion was fine. The shift seemed to happen right where the coin cell holders on one of the battery boxes ended, so maybe the nozzle got stuck on some filament blob? Any other ideas on what could have caused this?
One thing to note: I downgraded firmware from 5.0.0-RC to 5.0.0-alpha4 right before starting this print to improve WiFi speeds (the latest RC had a regression to slow WiFi speeds, which was fine in alpha4). Not sure if it's related.
I printed this at 0.2 mm structural with Polymaker PETG using the Generic PETG filament profile and MK4 input shaper printer profile.
I'm using a custom G10 magnetic sheet. It may be part of the issue, throwing off the input shaper resonance, but I've printed many things on it using the MK4 input shaper profile without issue.
RE:
I would suggest you insert the USB stick into a computer, copy the gcode files to it, then insert it in the printer.
Been reading too many scary things about the WiFi transfer corrupting Gcode files. I have my printer connected via Ethernet and still transfer takes FOREVER, so I am only using the USB copy method. You can connect a short USB extension cable to the printer's USB port to save the port from all the plugging/unplugging (that's cheaper to replace than a broken display board) See: https://www.printables.com/model/560183-usb-mount-under-prusa-enclosure for one idea to make this more convenient.
Keep in mind that corrupt gcode could cause a breakage beyond just layer shift - that's too risky for me - I have confidence the USB writing problems will get sorted out eventually.
RE: MK4 X Layer Shift
I did a file diff to compare the gcode from the flash drive (transferred over wifi using prusalink) and the original on my desktop, they were identical. Just in case, I changed to using a faster flash drive. I also cleaned the nozzle really good and updated the firmware and prusaslicer to the latest versions available. I've done about 3 similar prints since then and they've all worked perfectly... so hopefully it's fixed. Not sure the exact root cause.