Geometric Calibration + hotend?

[rampeh]

Hey guys,

I was inspired by the ABC tower rotation thread, and have spent the day painstakingly tweaking my rostock max. I whipped out my dial indicator and adjusted, checking 6 test points along a 110mm circle as well as the centre of the bed, aiming to get them all to have the same z level. I managed to get each of the test point within +-0.02mm of eachother with the dial indicator. I had to change rod length, delta radius, endstop heights and tower rotations. I was a bit surprised i had to change my rod length (269mm -> 274mm) even with the tricklaser carbon fiber arms.

I then remounted my hotend and did a quick paper drag test, and despite the dial indicator having great tolerance, the nozzle bore down too hard near the towers! Printing the bed levelling aid gives low points near the towers and slightly (much better) but slightly higher regions opposite towers.

Can you guys think of any reason that with the hotend the calibration would be significantly different? The two i can think of are:

1. The dial indicator is mounted above the carriage while my hotend is mounted below, giving about 50mm difference in printer height. I don't think my towers are crooked enough for this to be a major issue but i'm not sure.
2. The added weight of hotend + fans + load from the cabling causes missing steps which has some effect on the calibration.

Any ideas?

P.S I think i've learnt a lot about the error profiles for the different calibration settings; hopefully i'll have time to write up a little what i've learnt.