So what did I do wrong?

[nebbian]

Also try measuring the distance between the towers with a caliper. The actual distance between towers isn't important, but the measurement between top and bottom on the same face must be no more than 0.05mm different.

[Fattymcbiggins]

Grrr. I just checked and it somehow got changed to 7 factor. I've changed it to six and pushed an update.

g.

I don't know what the difference is between 6 factor and 7 factor calibration. But, the calibration tool linked from the Dozuki.com online manual page leads to a page where the result of the calibration process is based on 7 factors. Should I be looking somewhere else?

I have been trying to use this tool and seem to be getting results that are diverging.

Here is a snapshot of my latest result.

Success! Calibrated 7 factors using 10 points, deviation before 1.06 after 0.26

Enter these NEW values TO your EEProm
New Tower Endstop Offsets: X: 43 Y: 0 Z: 42

New Diagonal Rod Length: 304.15

New Horizontal Rod Radius: 150.10

New Z Max Length: 399.59

New Tower Alpha Angles: Alpha A: 210.44 Alpha B: 329.77 Alpha C: 90

[tjschweizer]

Interesting, I was getting a deviation of 0.04mm when using the Octoprint autocalibration script v0.10, then I saw this thread and updated to v0.14 and now the best I can get is 0.16mm.

[geneb]

The 7 factor calibration fiddles with the delta rod length, which can introduce scaling errors.

g.

[Zob]

OK, after trying a heap of permutations with arms & hot end orientation, all fruitless, I came across a posting in the forums that vaguely implied that running G29 multiple times might be beneficial.

I ran a G29 five times followed by two iterations of the OctoPrint auto-calibrate. Now reading 0.9, which is, for now acceptable.

More to the point I can print a solid first layer across much of the bed!

The model I was trying to print came out OK first time.* OK, it has a few snots and some light stringing but dimensionally it's bang on.

Thanks to everyone that tried to help. It seems I was just being too thorough by resetting the EEPROM between attempts.

Zob

*well, second time, I forgot to remove the z-height adjustment in the slicer the on first run <hears mhackney saying "I told you so" >