3
$\begingroup$

After approx. a year of more or less problem-free printing (40 days print time) with Prusa i3 mk3 kit, we started to suffer the random layer shifting on Y axis.

  • We inspected all the motor connectors are properly inserted in the mainboard and the wires are intact.
  • We ensured all the pulleys' screws on the x/y motor shafts are very tight and secure to be sure the pulleys don't slip.
  • We inspected the belts and pulleys, there is no mechanical problem like dust blobs, filament residues, etc.
  • We tightened the belts to be hard enough to show the expected support|belt status value (under 280) after self test is performed.
  • We switched off the crash detection feature.
  • We fully cleaned/re-lubricated the Y axis linear bearings and the rods to be sure there is no dirt causing any friction.
  • We tightened the nuts on the U holders fixing the bearings to the bed (hard enough not to squeeze the bearings so they can freely move on the rods).
  • We bought Japan PNY LM8UU bearings and replaced the original ones on Y axis (suspecting they are worn out).

Nothing of these helped fully:

  • Some steps improved the quality (like only one random skip in print).
  • Some steps degraded the quality (like additional step skips in previously flawless X axis).

Since the issues started, we had 4 hour reference design to be printed again and again using the same setup/filament etc. It became reference because its print started to exhibit the skips. We refused to slow down the print or perform any other workarounds, because we wanted to make the printer great again ;-)

What else should have been done to revert back to flawless prints without random X/Y step skipping and thus layer shifting?

$\endgroup$

1 Answer 1

2
$\begingroup$

Long story short: Try to replace the stock power supply.

  • We connected UPS between the printer supply and the wall outlet and the skips reduced to one/two per print.
  • We replaced the supply with random old used Mean Well SP-320 (zero care about the power panic feature, just plain 24V connected) and voilà: reference design passed.
  • Also few more complex designs after supply switch are still flawless, so we hope that was the problem and it's fixed now.
  • This Q&A post is to help others to skip the major frustration and save some time.
$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .