I have a heavily modified DiscoEasy 200 from Dagoma, see picture:

enter image description here

The heatbed is mounted on 4 springs, on top of the original plate. I made sure the X carriage was parallel to the table. I then tried to level the bed manually, with the 4 springs: I pushed the nozzle into each corner of the bed, and used the paper-gauge method to adjust the springs so that the bed would always be at the same distance of the nozzle. I then setup the Z-offset on the machine and tried to print a mainstream bed leveling test from thingiverse. I use a 4 points auto bed leveling.

And I have a systematic problem. Every time I try the bed leveling test, it seems the right side of the bed is lower than the left one. The front and rear right corners seem to be lower than on the left side. I tried to expand the springs of the front and rear right corners, but the auto bed leveling at the beginning of my prints re-introduces the problem.

The right side seems to be ~100 µm lower than what it should be. It's not a lot, but it's enough for prints to fail.

I can't figure out what the problem is. Do you have any idea? it's driving me crazy to not understand where the problem comes from.


2 Answers 2


Did you check if your bed is parallel to the gantry?

It can happen that after leveling one side and moving to the other the adjustment there puts the first side out of alignment - I level my bed by doing some couple turns

  • $\begingroup$ > Did you check if your bed is parallel to the gantry? Not to the gantry no, but the bed is parallel to the table (more or less, checked it with a caliper, but quickly). How do I check relatively to the gantry? I also do a couple of turns. Also, if I disable auto leveling, the leveling seems all right everywhere. $\endgroup$
    – JPFrancoia
    Mar 18, 2019 at 13:39
  • $\begingroup$ @JPFrancoia the hotend is a known length. if you level the bed, you actually try to make it parallel to the gantry. You should just repeat the process some times for left-right-front-back again and again, so the error gets reduced with each itteration $\endgroup$
    – Trish
    Mar 18, 2019 at 13:41
  • 1
    $\begingroup$ @JPFrancoia Though nice, leveling the bed with respect to the table is completely unnecessary unless the X axis is parallel to the table. The bed should be parallel to the X axis movement. $\endgroup$
    – 0scar
    Mar 18, 2019 at 13:52

As I understand your question:

  1. You manually go through a leveling process and have it set up well.
  2. You allow the printer to "auto-level".
  3. You print, and the right side is too low by about 0.1mm.

Is it possible that the filament (which is on the left side) pulls up on the extruder when the extruder is on the right side? A small tug could lift the extruder making it appear the same as if the bed were too low.

If this seems possible, check for the extruder being vertically loose or wobbly. I've had filament pull forces cause slight head offsets on a home-brew printer.


You must log in to answer this question.

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