Exchanged delta arms, calibration is now off ...

  • Hi,
    I finally exchanged my jury-rigged delta arms with magball ones, and after re-calibrating for different arm length (using the guide from the wiki), I now have the problem that the nozzle goes to zero trying to print the bed after a g32 calibration ... it looks like baby-stepping does not help (or does not help quickly enough). The filament can't leave the nozzle and the extruder starts stalling quickly (after using up the slack inherent in the bowden tube).

    I have attached a simple bed level test gcode I try to print


    And my config.g 0_1547661058068_config_g.txt

    Any ideas how to address this are very welcome (-:

  • @oliof Have you tried to adjust the trigger height on the G31 command in config.g (and config-override.g if that file has a G31 command?)

    Your config.g shows a positive value for the trigger height which suggests a Z probe that is mounted on the effector (such as a BLTouch, laser z probe, etc) that might have shifted when changing arms...

  • I adjusted the trigger height according to the instructions in the wiki. I will try to negate the value measured and see if that was it.

    (I am using the Duet smart effector)

  • @oliof said in Exchanged delta arms, calibration is now off ...:

    (I am using the Duet smart effector)

    If you're using the smart effector, then your Z trigger height should be a negative number (not positive) because the probe should only trigger AFTER the nozzle has pressed into the build plate. The G31 line in your config.g is:

    G31 P500 X0 Y0 Z0.313

    This is telling the duet board that when the smart effector triggers, the nozzle is 0.313mm above the build plate (and so hasn't made contact yet.) With this configuration, when the duet tries to move to Z=0, it will move until the z-probe triggers, and then move down an additional 0.313mm. That would probably be grinding the nozzle into the build plate.

    If the 0.313 number comes from testing with "G30 S-1" after testing for contact and then doing G92 Z0 (as outlined here:, then it's likely it should be a NEGATIVE 0.313 instead of a positive. In that case, your G31 line should appear as "G31 P500 X0 Y0 Z-0.313". In that case, when the duet tries to move to Z=0, it would move until the z-probe triggers, and then move UP 0.313mm.

    Good luck and please let us know how this works out

  • @garyd9 thanks, that was absolutely correct! Checking my config.bak the previous value was at -0.414, so I must have messed this up when adjustig the trigger value.

    The test gcode posted above looks perfect now.


Log in to reply