I tried the bearing/guide on the top of the Z rods and it made the problem much worse.

These two videos show the difference between ON/OFF heat bed vs PID.

ON/OFF

https://www.youtube.com/watch?v=0T5BE5XC2-8

PID

https://www.youtube.com/watch?v=RRSI2TJ3Trs

Those aren't my videos but some I came across. You have to figure with a 12" bed it would be even worse.

Quote Originally Posted by lakester View Post
idle speculation...

(FWIW..., the one mod I was contemplating was the addition of the bearing/guide for the top end of the z-rods..., haven't tried it yet)

..., back to speculation...

I wonder if a firmware mod might be worth a shot, specifcally, one that tweaks the "hysteresis" assumptions regarding bed heating.

My intuition doubts that the periodicity of heating cycles substantially change the overall expansion/contraction of the glass or overall structure in a fashion that would correlate with the z-banding issue. The heating system is more or less responding to the behaviour of the portion of the system with the least thermal mass, i.e., the heating pad itself. Is it really plausible that the overall structure responds significantly with the same periodicity?

I'm guessing that heat cycling correlates poorly with what is essentially a geometric effect in the form of z-banding.

Just the same..., maybe it would be worth reducing the hysteresis built into the heating algorithms and see if it changes anything wrt z-banding.

Again..., my feeling is that there is a geometric problem at work here, independent of heating..., but maybe if the heat cycling were tweaked..., it would help settle the issue once and for all.

0.02USD

(additional thought: heck..., for the sake of experiment, what if we disabled heating altogether, and did a z-banding test print with PLA on blue tape?)