Printer Model: S1 PRO
MCU / Printerboard: R Pi klippy.log Unconfirmed 832850.txt (986.3 KB)
Fill out above information andin all cases attach yourklippy.logfile (use zip to compress it, if too big). Pasting yourprinter.cfgis not needed Be sure to check our “Knowledge Base” Category first. Most relevant items, e.g. error messages, are covered there
Describe your issue:
…
Good day everyone. This is my first 3d printer and installed Klipper after playing with Marlin for a week. I’ve had Klipper running for 2 days and am figuring things out and really liking all the features it offers. I found last night the bed mesh viewer. very cool, and noticed a “bump” in the image. I tried the following things to determine where the anomaly is originating but no matter what I do, it remains. I rotated the PEI sheet, removed the PEI sheet, then did both again with a hot bed 65*. Max deviation is .213mm.
Anybody else have similar results and is this anything to worry about? Thoughts on how to correct it?
Yes, I pulled the PEI sheet and looked for any little particles, nothing, but I also ran the bed mesh without the sheet, with it rotated and it made no difference, the profile looks pretty much the same for all scenarios.
Okay thank you. I modded the print.cfg file for a 16x12 bed mesh and ran it, immediately something jumped out at me, there are now 3 lumps, front to back back along the Y axis. I thought about it, bed width is 220mm, the spacing of the lump peaks are exactly 71mm apart for all three, the diameter of the rollers where they ride are 22mm, the circumference of the rollers is 71mm…Mmmmm, there is a roller that is not quite round as it’s not a sharp peak, it ramps up to a peak so it won’t be a pebble or xxx, I’ll try cleaning first, any other thoughts?
Thanks for asking me to refine the bed mesh, cool results.
Yep, sounds like your culprit. Wiping the rollers and the gantry surface down is worth trying. You can also try to manually turn one of the top rollers by hand about 180 degrees so that it is out of “phase” with the other roller, just in case the two have developed a uniform deformation. If so, then rotating one roller and not the other can sometimes help to make the effect of the deformation less pronounced.
Either way, unless cleaning resolves the problem, it’s probably worth getting some replacement delrin rollers.
30 seconds after I posted my son suggested turning 1 roller at a time about 45*, then test again, we did that 3 times and it’s the bottom roller, the top two made no difference. The printer works perfectly, is this something that should be resolved or I am chasing a measurement “just because I can”? I will go through and clean everything, it’s only a week old but could be something from the factory. Are Delrin rollers upgrades from stock?
Thanks for your help.
PS while I have your attention, is there a program, macro, file or ?? that will allow me to adjust the corner screws similar to the way the stock Creality/Marlin allows to to “paper test” in 5 spots on the bed but in Klipper? I don’t want to reflash Marlin just to do a mechanical bed level. thx
Well this is the best I can get the bed level, ignoring the bumps we are discussing, I’m getting a max deviation of .145mm, I think that is 145 microns? I read somewhere we should be shooting for 25 microns max, is that correct? Going to start looking for new rollers.
Yeah, I would try to address that. 0.145mm is nearly 75% of the “standard” layer height of 0.2mm so the max deviation is almost the thickness of a “standard” layer. Also, the bed mesh is intended to compensate for a build surface that isn’t perfectly flat, not to compensate for mechanical flaws in the printer’s motion system. Yes, it can help partially conceal the issue with the deformed roller but can result in reduced dimensional accuracy and consistency.
Excellent point! In this regard it would be a similar phenomenon to the gantry twist issue insofar as calculating a mesh with a problematic motion system actually amplifies the problem.
In essence, the “spikes” on your mesh will actually be translated in the X coordinate according to your probe offset since they were caused by the motion system and not the bed itself. This may result in two sets of problem spots: One wherever the flat spots on the roller actually are, and the other where where the mesh erroneously thinks they are due to the probe offset.
