Dramatic layer shift on X
Hi all,
recently having an issue where prints are quite dramatically shifting along the X axis on one layer only.
Not really sure where to start on this since calibrations otherwise seem good and consistent. Any advice is appreciated!
This is printed with prusa petg. Based inside an enclosure, moisture fine, filament dried in a heater.
How to diagnose layer shifts
Since this is X, I'll start there - there are 3 components involved. Pully Tightness. Belt tension (vs drag), and Bearing squeeze on the X carriage back screws.
I recommend checking and remediating all 3 if you find gaps.
Pulley screws - the order these get tightened in matter. Tighten the screw on the pulley shaft flat first. Then tighten the other one.
Belt tension. Stop looking at the #'s in the control panel, get yourself some prusament PETG, and print then use the belt tension gauge. If you use other filaments to print it, throw it out - since it won't read accurately. Adjust to the lower end of the sweet spot.
X carriage back bearing tension screws. Loosen them. Don't even check them. Just loosen them each by 3 turns. Then, for the ones on the top, check for "play" along the Y direction - and slowly tighten each of the two top tension screws a quarter turn until the play is gone. Do not tighten further. The screws on the bottom should be allowing some play in the Y direction on the bottom bar bearings.
I would add a fourth possible culprit: The nozzle hitting some crap in the infill. Could be the infill itself; some types have crossing lines within one layer and pileups at the intersections. Unlikely in your case, looks like gyroid. Or it could be dripping blobs of filament, not uncommon for PETG. Make sure your nozzle is properly assembled (no filament oozing from above the heater block. Nozzle and heater block are squeaky clean before the print. And perhaps bring the temperature down a bit.
I guess another remote possibility—but easy to check— is the heater cartridge, thermistor, or wire bundle from the extruder not properly position, so hitting the model.
I guess between that and the three factors Dan mentioned, not much else left.
-r
Brilliant, thank you both. I'm going to try these suggestions tonight and will follow up. I have a slight suspicion that it might be Fuchsr's suggestion, as I've had fairly consistent problems with a small part of the bed failing to adhere properly on first run. Here's an example of a successful print that still shows that artifact, that (thinking about it) did work between these two failed prints that I showed at the top of the thread.
You can see the weird initial bed adhesion on the upper left strut of the print.
