Bit of help with G32's and G29's



  • HI all,

    I'm a little confused with calibrating a delta.

    In my start script, I have it run a G32 delta calibration, and then G29 S1, to load the mesh level.

    Does a G32 also set the correct z-height automatically, or do I also need to do a G30 ?

    Reason I ask is I often have to play with baby stepping (an awful lot) to get sensible first layer adhesion.

    My result of a G32 is usually 0.032 so I think I'm within reasonable tolerances in terms of the build.

    Also, sometimes if I run G29, my result comes back with the expected low and high spots, but the mean/median position is the grid plane on the visualisation. Other times I run G29, the bed in its entirety is all above the grid plane…

    What is the correct procedure/timing for when to run G29?

    Thanks, Ben





  • Aha…well that explains the G29 overall offset then......so next open question is does a G32 automatically set correct z-height?


  • administrators

    G32 automatically sets the correct Z height; but if you subsequently run G28 then the Z height will only remain correct if the endstop switches trigger at exactly the same carriage heights as they did when you ran G28 before G32. If the temperature has changed significantly then that won't be the case, due to different rates of expansion of aluminium extrusions, belts, and carbon fibre arms.

    So between running G28 and the print actually starting, I recommend you do one of the following:

    1. Run G32 auto calibration.
    2. Do a single G30 probe from a few mm above bed centre.
    3. Run reduced calibration, e.g. 3-point calibration.



  • Hi David,

    Thanks for the input.

    My process (in start g-code) is usually

    1. G28 ; home
    2. Heat up extruder to printing temp, heat up bed to printing temp.
    3. G32 ; (runs bed.g, which does somtehing like 8 perimeter probes and 5 intermediate probes and centre probe)
    4. At the end of my bed.g file, I have G29 S1 to load mesh
    5. Run purgenozzle.g

    So in essence I'm auto calibrating at printing temperatures and printing begins with no homing move between the calibration and the print starting.

    Does that sound like I'm doing everything correctly?

    I did a big print (first layer 0.15) that took up almost the whole bed (240mm diameter) and was getting very squashed, almost non-existent extrusion on the left side of the bed, and much taller extrusion on the other side (calibration result was 0.039)


  • administrators

    That sequence sounds good to me.

    What type of Z probe are you using? The symptom you describe suggests that you are using a Z probe that is offset from the nozzle, and the effector tilt is varying with XY position.



  • No, i'm using a smart effector so that shouldn't be the issue.


  • administrators

    What does the mesh height map look like?




  • administrators

    That looks fairly flat apart from a few points right at the edges. When you run G32, how many factors do you calibrate?



  • 6 factor calibration - I've set the rod lengths myself as I had the calibration numbers from the labels on the mag-ball arms.


  • administrators

    In that case, I am puzzled!

    You could check that the trigger height (which should be around -0.1mm) is the same at both sides of the bed.


Locked
 

Looks like your connection to Duet3D was lost, please wait while we try to reconnect.