This is my printer:
1 Like
Very cool build.
Not what I ever would like to have as a 3D printer but very cool design 
2 Likes
Looks like vibration although the mechanic is quite slow.
But nice build. Could print endlessly if you keep feeding the filament pellets.
Thanks , for your answer , but the thing is that it is always on the same track , if it would be because of the vibration it wouldnāt be on the very same spot each second time ( first the printhead is going forward then backwards )
When it is printing in vase mode without changing direction there is no sign of backlash
Unfortunately, I have zero experience with this type of kinematics and I also do not know how well tested it is in Klipper.
In fact, you are the first report on this kinematics I have seen so far.
Interesting printer.
If I had to guess, Iād guess that you are hitting the practical resolution of the gantry rotation step precision. That is, Iād guess the gantry needs to rotate a very tiny amount to get the desired straight line, and there is not enough resolution to move that tiny amount (due to stepper microstepping limits, the gantry rotation gearing ratio, some kind of physical āstictionā in the motion, or something similar).
The history of Klipperās polar support is at Polar Kinematics Ā· Issue #578 Ā· Klipper3d/klipper Ā· GitHub . You could try reaching out to Thomas Herrman (BKLronin) on github to see if he has insight.
Cheers,
-Kevin
1 Like
Thanks, it might be an issue, although I tried from 16 to 128 microstepping and it was the same result, every second layer is on the right place. The gear ratio on the turning gear is 324 :18 (stepper) so the movement should be pretty smoth, and the other interesting part is that after some distance it compensates and the line ends in the right place just the beginning has an offset after the motor switching direction.
Iāve got the same problem with my ānormalā polar printer, when printing too fast / under too high accelerations. In my opinion it is highly probably a mechanical problem, not a software one. You are moving a loooot of mass there. Did you already try to print very slow and with low accelerations?
Hi Neelix, in our case the issue is very delicate, because when I reduced the speed to 20mm/sec and I turned the square in 45 degrees it is still the same double wall effect on two sides, but when I print a cylindrical object it is almost unnoticeable. It seems to me that there is some kind of problem with the interpretation of the slicer g-code through the klipper. https://youtu.be/xuN_btz5sR4
weiiiird. But the normal polar kinematics should be fine, since they work for many other printers perfectly fine