Massive layer shift after ~24 hrs of flawless printing
I've never had a layer shift like this. The angled direction makes me believe it's one of the XY motors. I fixed a loose pulley and corrected belt tension before starting this print. After the print, both motor pulleys look good and tension measured correctly. I'm stumped.
RE: Massive layer shift after ~24 hrs of flawless printing
Did you check if the build plate is really free to move on the low end? Could be one or both of the Z motors skipped due to high resistance, causing a collision.
Such problems are more likely to occur close to the motors, because the threaded rods cannot flex enough to compensate. For such high prints I made it a habit to drive the Z axis to the extremes before starting, listening to weird noises. On the XL specifically, you can look at the tops of the rods / bearing mounts. They should not move while driving over the whole range.
RE: Massive layer shift after ~24 hrs of flawless printing
Thanks for the help! I messaged Prusa, Andres and I did some diagnostics and over several prints noticed a reliable failure: the small diameter rods which extend at an angle will break off. I even tried the XL version of the rocket engine. This print was originally intended for 0.6 mm nozzles. When I printed it I tried 0.4 mm, 3 perimeters, and 20% infill. This seemed to provide a substantial amount of material through that cross section, but it seems it's not enough. Andres attempted the same print and it failed in the same way.
As a Prusa test print, I'd like to arrive at a reliable method to manufacture this, but it seems it won't without a substantial amount of trial and error, which is expensive.
RE:
Seems weird to me. They are showing off this print, so it should print, right?
I'd rather think it's a slicing issue, or an issue with the newer firmwares, or his printer has the same problem as yours 😉