1. Got a question or need help troubleshooting? Post to the troubleshooting forum or Search the forums!

Bed Levelling issues

Discussion in 'Troubleshooting' started by J.P., Jun 1, 2018.

  1. J.P.

    J.P. New Member

    Joined:
    Jul 11, 2017
    Messages:
    8
    Likes Received:
    0
    I'm curious to know if anyone else has had leveling problems with the R2.

    So the symptom i have is the first layer is trying to print too close to the bed so no filament is coming out and the extruder is clicking. I'm still using the original Glass bed with the POGO board. I'll list below all the things I have tried to do to solve the problem.

    I have tried running the two bed leveling wizards multiple times, always with the same result (see above)
    I have tried to re-slice the project
    I have tried a known successful project from another printer of the same make and model
    I have tried to adjust the thumb screws during the print so it will extrude properly, once its printing properly I have stopped the print job, taken the partial print off, then resent the same job to the printer to have the symptom reappear.
    I have tried reformatting the memory card for the RaspberryPi and putting a known working image on it, updated it back to the current version, relevel the bed, reslice a known good print job, and the symptom is still there.
    I have made sure the print bed is completely cleaned and clear of any objects, including glue.

    All jobs sent to the printer are the exact files and gcodes that have worked on the exact same kind of printer so I know the file should print. At this point I think it is either the auto leveling sensor or the print bed itself. Has anyone else had this issue?
     
  2. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    Try manually adjusting the Z offset up/down and see if it makes a difference. Forget the wizards, just try editing it a smidgeon and see if it makes a change.

    It is possible the IR sensor for the bed position is not reporting correctly.
     
    Geof likes this.
  3. drbanks

    drbanks Active Member

    Joined:
    Dec 29, 2017
    Messages:
    141
    Likes Received:
    40
    As much as I try to be precise with the Z-offset and fine tune wizards, it still takes me two or three prints before I really feel that I have it dialed in. Usually, after each of the first few prints after the wizards, I'll use the EEPROM editor to tweak the Z offset up or down by .05mm until I have the right "squish" and first layer texture. As a safety, until it's dialed in, I'll use rafts to ensure that I get something useful out of the printer even if it isn't perfect.
     
  4. drbanks

    drbanks Active Member

    Joined:
    Dec 29, 2017
    Messages:
    141
    Likes Received:
    40
    A Week or two after upgrading my bed surface with a flexiplate system, I started getting probe crashes during the levelling process. I just figured at the time that maybe I wasn't aligning the plate with the magnetic bed well enough and after a few tries to get it to work, just commented it out of my Octoprint startup script. I've been working that way since.

    Over the weekend, I was getting some clicking out of the extruder on the first layer, and on investigation discovered that the bed screws had rattled loose over the last few weeks. So, starting at the one corner that was still getting a healthy line, I re-adjusted the other three to get the right height.

    But, I started to worry that the last good probe data from the bed levelling might no longer be accurate. So, I've been doing some debugging to see where the problem is.

    The short answer is I think my BuildTak surface is too dirty. Funny because I ran the original non-flexiplate BuildTak surface for months without an issue. Until I printed some PETG without a release agent and ended up tearing half the BuildTak off the bed, which is why I decided it was such a good idea to do the Flexiplate upgrade just then. (It'd been sitting in the box for weeks).

    The new BuildTak sheet for the flexiplate really looks/looked a lot cleaner than the old EOM sheet. The older one was covered with gluestick and various holes from extruder crashes. I didn't even know that the bed was dirty (from looking at it) until I felt the bed and discovered that it had a fairly thick layer of hair spray on it. Which, btw, doesn't come off so easily.

    But, shortly after I installed the system, I went back to the BuildTak site and ordered a few more sheets just to have handy. At the time, I also ordered another removable flex plate and a sheet of PEI. I liked the PEI at first until I had a hot-end destroying sudden loss of bed adhesion on my new Prusa MK3 and learned that you ABSOLUTELY have to clean the bed with IPA at the beginning of every print. And, for that matter, watching the first two layers go down on PEI isn't good enough, because I was getting separation around layer 10.

    Anywho, I'm watching my R2 nozzle crash trying to do the levelling routine, over and over always in the front row in one of the first three positions. It's about this point I'm wondering how much life you can expect to get out of a hardened nozzle if you keep letting the nozzle crash like that. I was cleaning everything in and around the extruder that I could think of and trying to guess exactly where the probe was, but after a few paper towels and way too much IPA, I decided I was on the wrong tracik.

    Out of frustration, I put the PEI plate on instead, and first time, every time the bed levelling works without crashing.

    so, trying to decide whether to re-enable the levelling in the Octoprint gcode script, or if I should do it manually any time I have to readjust a bed thumbscrew or the Z-Offset.

    And, I'm getting kinda jiggy with the PEI again, and I guess I need the practice for when the replacement Prusa shows up tomorrow.
     

Share This Page