Auto leveling woes

Discussion in 'Calibration, Help, and Troubleshooting' started by nemith, Mar 9, 2016.

  1. nemith

    nemith Member

    Joined:
    Mar 9, 2016
    Messages:
    16
    Likes Received:
    14
    I have had nothing but pain when it comes to the Z-axis limit sensor and auto leveling. When printing on just bare glass it seems to work ok, however prints have a very difficult time sticking.

    I have decided to try PET tape with glue on top for PLA or ABS Juise for ABS to keep the corners down. The auto level sensor with PET on the bed seems to just freak out. I have had the autoleveler crash the bed many times.

    Anyone have any tips for how to get the z-axis sensor working reliably?
     
  2. Madeinoz67

    Madeinoz67 Active Member

    Joined:
    Sep 8, 2015
    Messages:
    38
    Likes Received:
    12
    A gluestick always seems to work for me when I have problems not sticking to the bed.
     
  3. UlrichKliegis

    UlrichKliegis Well-Known Member

    Joined:
    Sep 5, 2015
    Messages:
    187
    Likes Received:
    47
    I prefer a glass of red wine or a beer.
     
    mike01hu, Madeinoz67 and PsyVision like this.
  4. UlrichKliegis

    UlrichKliegis Well-Known Member

    Joined:
    Sep 5, 2015
    Messages:
    187
    Likes Received:
    47
    My FSR based solution (description in german, english explanation here) might be a bit harder to implement in the BB, but I'll think about that. Meawhile, you might look at this.
     
  5. Ric Klaren

    Ric Klaren Member

    Joined:
    Sep 4, 2015
    Messages:
    6
    Likes Received:
    1
    I found this issue on the Marlin firmware that describes the trials someone else had with bedleveling.

    https://github.com/MarlinFirmware/Marlin/issues/1244

    Want to play around with the various tips and instructions in there. But it seems my holidays will prevent me from playing with the BigBox ;) What I get from the issue is that it's very important to verify the Z is moving smoothly. But I somehow suspect te sensor is a bit finicky with the glass and whatever is on top of it. Something to try out is to but some paper on the corners that are used for the probing, that might make things react more consistently, that only leaves the initial autohome though. One could suspect that the autohome is used for an initial offset and then the bed leveling becomes leading in what the printer thinks of what the Z should be (pure speculation)

    See my most on Z crashing after bed leveling for things I already tried out. My experiment with paper on top of the bed seems to indicate the sensor is inconsistent on the plain glass (and glass with glue). Or it might be related to how light of the surroundings falls on to the glass.

    Cheers,

    Ric
     
  6. UlrichKliegis

    UlrichKliegis Well-Known Member

    Joined:
    Sep 5, 2015
    Messages:
    187
    Likes Received:
    47
    Did you consider to cover the BB with a dark cloth to just verify or exclude that assumption? What kind of illumination do you have around there?
     
  7. mike01hu

    mike01hu Well-Known Member

    Joined:
    Apr 22, 2014
    Messages:
    957
    Likes Received:
    166
    That's why I have so many empty bottles!
    Mike
     
  8. nemith

    nemith Member

    Joined:
    Mar 9, 2016
    Messages:
    16
    Likes Received:
    14
    It seems to be definitely related to how you coat your bed. The IR is super finiky.

    If i use a mixture of elmers glue and water i can get it to crash regularly. Same with just PET tape. Glue stick seems ok, but i don't get even first layer. Glue stick and ABS doesn't work that well however for large flat pieces and I really want to do PET + ABS Juice, but i cannot even get close to getting the bed level enough to pull it off.

    I have just a regular overhead light off to the side of my printer.

    I have been experimenting with using a piece of thin cardboard paper to do the auto home / leveling and then removing it before it prints, but the cardboard paper is flat at room temperature but on a hot bet it starts to curl (but i am still way closer than NOT using it).

    The sensor seems just really unreliable. I wonder how hard it would be to move to a mechanical probe (like in the early videos of the prototypes). As it is springs + manual leveling seems like a better option to get a good even first layer.
     
  9. Ric Klaren

    Ric Klaren Member

    Joined:
    Sep 4, 2015
    Messages:
    6
    Likes Received:
    1
    Didn't have time yet to verify, good suggestion though. Will go on holidays soon so suspect this will come after those.

    Thanks,

    Ric
     
  10. nemith

    nemith Member

    Joined:
    Mar 9, 2016
    Messages:
    16
    Likes Received:
    14
    I've decided to try Manual/Mesh leveling and so far I am liking it better than the auto leveling madness. I am reapplying some PET tape to another glass bed that i bought to try that with the mesh leveling.
     
  11. nemith

    nemith Member

    Joined:
    Mar 9, 2016
    Messages:
    16
    Likes Received:
    14
    Mesh leveling is so much better imho.

    This is on PET with ABS and so far the first layer is looking great. [​IMG]
     
    Mike Kelly likes this.
  12. elmoret

    elmoret Administrator

    Joined:
    Mar 1, 2014
    Messages:
    586
    Likes Received:
    91
    Reading the documentation for the sensor might save you some headaches:

    https://miscsolutions.wordpress.com/mini-height-sensor-board/

    The creator of the sensor, David Crocker, is a really helpful guy as well. I'm sure he'd be happy to help if you drop him a line.
     

Share This Page