Autocalibration bias related to what factors?
Posted by sungod3k
|
Autocalibration bias related to what factors? June 01, 2017 06:37PM |
Hi,
Im doing well with smoothie and Re-arm and I spent some time today to square up my frame. Its a big and slightly fragile frame: 1m high and 40cm triangle sides, but with some cross-bracing and flying extruder I have no instabilities.
I tried to get all frame related variables as close as possible. Arm length, arm parallelity, triangle sides and height are +-0.5mm.
For auto calibration I ran G32 and G31 to 0,02mm reliability but after some test print one half of the print is a bit too high, a pattern that I've had before but much worse.
My squaring up of the frame definitely helped but if I want to use my full 30cm diameter bed the current auto calibration would not serve the whole bed well.
I also repeated the test print without G31 and came to the same result, the bed is pretty flat anyway.
I also ran a G29 and the are close enough for my taste. The "delta max" reported was 0.11 but I couldn't find what that actually measures.
What I can see from the high spots is that it run perpendicular to an axis from the beta tower to the middle and I will try to manually correct the beta endstop tmr.
Now Im curious if someone has an idea which part in the frame I should try to get more precise or which factor the auto calibration relies on the most/throws it off the easiest.
Cheers
Im doing well with smoothie and Re-arm and I spent some time today to square up my frame. Its a big and slightly fragile frame: 1m high and 40cm triangle sides, but with some cross-bracing and flying extruder I have no instabilities.
I tried to get all frame related variables as close as possible. Arm length, arm parallelity, triangle sides and height are +-0.5mm.
For auto calibration I ran G32 and G31 to 0,02mm reliability but after some test print one half of the print is a bit too high, a pattern that I've had before but much worse.
My squaring up of the frame definitely helped but if I want to use my full 30cm diameter bed the current auto calibration would not serve the whole bed well.
I also repeated the test print without G31 and came to the same result, the bed is pretty flat anyway.
I also ran a G29 and the are close enough for my taste. The "delta max" reported was 0.11 but I couldn't find what that actually measures.
What I can see from the high spots is that it run perpendicular to an axis from the beta tower to the middle and I will try to manually correct the beta endstop tmr.
Now Im curious if someone has an idea which part in the frame I should try to get more precise or which factor the auto calibration relies on the most/throws it off the easiest.
Cheers
Sorry, only registered users may post in this forum.