Bad height map?



  • This thing looks way wonkey. I am using an aluminum heat spreader, then glass, then PEI with black paint baked onto the pei. why does it look so bad.?


  • administrators

    The spikes at the top right suggest to me that you have reached the movement limit of the joints in that region. The general red area top left suggests to me that you have have varying effector tilt. I have a bulls-eye spirit level on top of my effector so that I can see tilt.



  • Thanks.

    I did get a bullseye level, but it was 60mm so too big, I got a 15mm coming.

    Once I verify the tilt, what can I do about it?

    Also, I ran g29 after doing the auto calibration with tilt adjustment. I thought I read that the height map doesn't take it to account the effector tilt adjustments, or is this indicative of bad-adjustment?


  • administrators

    With G32 you have the option of adding H parameters to the G30 commands in bed.g to correct for differences in trigger height caused by effector tilt. With G29 that's not practical because of the much larger number of probe points in general. So G29 is only really usable on a delta if you have little or no effector tilt, or a nozzle-based probe that is insensitive to effector tilt - although in that case, effector tilt will still give rise to print distortions.

    For causes of effector tilt, see the first section of https://duet3d.com/wiki/Calibrating_a_delta_printer.



  • Thank you, I will go through that.

    In the mean time, are you saying that using grid-based compensation and auto calibration will produce worse results than just using autocalibration alone for a printer with effector tilt?



  • Assuming I have a valid method for testing, it seems like the dumb-bells on the seemecnc trucks are not level with the bed.
    @Duet WifiWiki:

    The two bearings on each carriage must be at the same height
    The bearings in each pair on the effector should be at the same height by design. The bearings on the carriage at the other end of the rods must also be at the same height, otherwise the effector will be forced to tilt when it is close to the tower with that carriage on it. Carriages are typically screwed to carriage trucks or linear rail sliders, and you can usually rotate a carriage slightly about the axis passing through its face to get the bearings at the same height before you tighten the screws.

    Because the dumb-bells are not level, the bearings (or balls) are not at the same height, and this is causing? effector tilt. I took some pictures to better explain.

    Here is the X axis. The first picture shows the bed, and the level is oriented the same way as the trucks are. The bed is level, but the dumb-bells lean a little bit (i also removed the arms and tested (not pictured) but the result is the same).

    X axis at bed:

    X axis at dumbells:

    I also placed a level on the tower to see if the tower was leaning

    Here is the Y axis:

    Here is the Z axis:

    Sadly, there doesn't seem to be a way to adjust this. Should i look at getting new trucks?


  • administrators

    I don't know how the seemecnc carriages are constructed. On most delta printer designs i have seen, the carriages are screwed on to the sliders or trucks; so if the mounting holes on the carriages are a little oversized then you can rotate the carriages on the sliders or trucks a little before tightening the screws.

    Perhaps you should ask for suggestions on the seemecnc forum.


Log in to reply