SOLVED Unable to Mesh Bed Level satisfactorily

Discussion in 'Calibration, Help, and Troubleshooting' started by ndpg, Jul 28, 2016.

  1. ndpg

    ndpg Active Member

    Joined:
    Apr 20, 2016
    Messages:
    31
    Likes Received:
    23
    The consensus seems to be that MBL is the way to go so I installed the appropriate firmware for a Hybrid Dual (Titan), Now I am struggling to level satisfactorily.
    When I set the Z datum with the strip of belt the nozzles are just above the bed with Z displayed as zero. Makes sense since that would be good enough for ABL to work.
    However, when I try to MBL the bed backs off 4 mm before each of the 9 sample points. The problem is that MBL only allows 4 mm of adjustment so I can never reach the point where the paper is getting tight since it only takes me back to where I started. Why back off to only the limit of adjustment? Surely it should back off by say 2mm so I had +/- 2mm to work with.
    My Z offset is 0. Should it be something else?
    I'd like to think that I'm doing everything as described in the instructions but they are so inconsistent that it is hard to be sure.
    My logic is this:
    Setting the IR sensor provides a zero datum for Z at some distance just above the bed. That distance is somewhat arbitrary since it is described as being 'just above' rather than an actual distance.
    The problem is that the setting is made at one point and the actual bed height will vary away from that point. ABL or MBL are intended to fix that. ABL will compensate for variation but only with respect to the original gap between head and bed. MBL will compensate based on a specific gap (the thickness of the piece of paper) and at more locations, which is why MBL should be better.
    To repeat my question in a different way: How do i get MBL to measure either side of the Z datum rather than just one side of it? (Or, how do I get it to work sensibly?)
     
  2. PsyVision

    PsyVision Moderator
    Staff Member

    Joined:
    Sep 4, 2015
    Messages:
    798
    Likes Received:
    248
    MBL shouldnt restrict the amount that you can adjust the Z for each of the 9 points. The display may show 4mm which reduces to 0mm but you can keep winding it and go further, its more of an issue in how the Z displays on the LCD. Is that what you are seeing?
     
  3. ndpg

    ndpg Active Member

    Joined:
    Apr 20, 2016
    Messages:
    31
    Likes Received:
    23
    It does restrict me. I get to 4 mm and can't go any further so no matter what I can't get the nozzle low enough to grip the paper. Also it starts at zero not 4 mm so I have to increase it not reduce it. Surely MBL is a fine tuning operation, so why does it start so far away (4 mm) from the nozzle? It's as though it is 4 mm away from where it thinks it is.
     
    #3 ndpg, Jul 28, 2016
    Last edited: Jul 28, 2016
  4. Alex9779

    Alex9779 Moderator
    Staff Member

    Joined:
    Sep 4, 2015
    Messages:
    2,405
    Likes Received:
    731
    @ndpg which firmware did you download and when?
     
  5. ndpg

    ndpg Active Member

    Joined:
    Apr 20, 2016
    Messages:
    31
    Likes Received:
    23
    Yours I believe. Marlin.RC_Hybrid_Titan_MBL.hex built on July 9th according to the page on Jenkins.
     
  6. Alex9779

    Alex9779 Moderator
    Staff Member

    Joined:
    Sep 4, 2015
    Messages:
    2,405
    Likes Received:
    731
    Hmmmmmm strange... Will test it and see how it reacts...
     
  7. Alex9779

    Alex9779 Moderator
    Staff Member

    Joined:
    Sep 4, 2015
    Messages:
    2,405
    Likes Received:
    731
    @ndpg ok i tested and it behaves as expected, I have no unusual rain, the nozzle is within reach of the adjustment of +-4mm...

    If you load failsafe defaults and home the head how far are your nozzles away from the bed?
     
  8. ndpg

    ndpg Active Member

    Joined:
    Apr 20, 2016
    Messages:
    31
    Likes Received:
    23
    After the failsafe and a home the nozzles are about 4 mm from the bed and the Z offset is 5 mm (it was 0 mm). After MBL the nozzles home to the paper width above the bed. I just able to MBL but at two locations I reached -4.000 mm. The others were -3.625 mm or more. The IR sensor was set with the piece of belt as in the instructions. That doesn't set where it 'sees' the bed is though. Maybe I should back it off a bit so the nozzles end up closer to the bed.
     
  9. Stian Indal Haugseth

    Stian Indal Haugseth Well-Known Member

    Joined:
    Sep 11, 2015
    Messages:
    573
    Likes Received:
    95
    Maybe you need to raise the IR or tweak it. The IR diode and photo diode might be unaligned.

    Good to know of the 4mm limit :)
     
  10. PsyVision

    PsyVision Moderator
    Staff Member

    Joined:
    Sep 4, 2015
    Messages:
    798
    Likes Received:
    248
    Can you video yourself doing the process and maybe post a photo of the IR position? Neither Alex nor I have any problems with limits during the MBL process.
     
  11. Alex9779

    Alex9779 Moderator
    Staff Member

    Joined:
    Sep 4, 2015
    Messages:
    2,405
    Likes Received:
    731
    Actually I had, but was my own fault, I change MESH_HOME_SEARCH_Z to 1 in a test config and the the limit was +-1mm :D.... In all the firmwares published it should be +-4mm...
     
  12. ndpg

    ndpg Active Member

    Joined:
    Apr 20, 2016
    Messages:
    31
    Likes Received:
    23
    I set the IR sensor with two bits of belt with the teeth engaged. That's about 0.9 mm. Now my MBL offsets are all -3.5 to -3.6 mm. It seems that I do have a sensor that is a bit longer range than normal (or thin belt!). So long as it is accurate that shouldn't matter.
     
    Stian Indal Haugseth likes this.

Share This Page