I'm working on scripts and a procedure to calibrate Marlin based Deltabots.

3D Printing
,
1

I’m working on scripts and a procedure to calibrate Marlin based Deltabots. Code is available. This is still a work in progress.

It seems to me like there 4 main parameters which are “tweaked” in a typical deltabot calibration. The DELTA_RADIUS, and the offsets on the 3 tower endstops. I find errors for these parameters which best describe the measured values. These errors can then be subtracted from your settings for a fast calibration convergence.

This is a work in progress, and I will add user documentation as it progresses.
https://github.com/Mr-What/Marlin/tree/deltabot/util

2

Can’t wait for documentation :slight_smile:

3

Looks great. Haven’t tried it yet myself, but have you seen what Rich Cattell has done with autocalibration in the Marlin firmware itself ? https://github.com/RichCattell/Marlin

4

I’ll look at that firmware. It does far more than I do at this time. My technique is kind of a sledgehammer. Not practical in firmware. He may have some more elegant mathematics that I should look into.

5

@Andrea_Fontana Keep on my case. At this time I cut and paste output on pronterface into a perl script that extracts (x,y,z) calibration points, which should be on a z=const plane. This data is fed into an optimization routine that guesses the calibration errors, in octave. I need to look at RickCattell’s work to see if he has done everything I wanted to do, but more and better.