Printer Model: Modified Ender 3 Pro
MCU / Printerboard: SKR 1.4 / Raspberry Pi klippy.log klippy.log (1013.6 KB)
Describe your issue:
Hi!
I have a problem with an “indent” before every corner and slightly overshoot after every corner.
What i did to hopefully solve the problem:
different belt tensions
an different extruder
new TMC2209 drivers on every axis
changed square corner velocity
changed max_acceleration
changed max_accel_to_decel
with or without PA
with or without Resonance Compensation.
No change at all.
BUT. When i do the resonance compensation tuning tower with all the commands in the manual the corners are good - no “indent” or underextrusion before corners and almost no overextrusion after corner.
maybe it ihas something to do with unsynchronized extruder and x/y motion?
tackled this for a very long time now. i dont know what to do.
Is x and y running smoothly? Yes. Cleaned all the linear rails and blocks and greased them.
Extruder OK not full of dirt? Yes. No dirt. I even used my old Titan Extruder instead of the Orbiter V2.0 - same deal
Extruder gears correctly running and smooth? I think so. The Orbiter V2.0 is almost brandnew. The Titan Extruder is old but back in the days (before some Klipper updates i think or even before Klipper) everything was fine and no dent before corners.
Nozzle clogged? I recently printed with a new one (0.4 instead of 0.6) - same deal
Will recheck everything. Thank you for your help
Edit: I think its a software problem. Even when i rotate the printing object by for example 45 degrees the dent is exactly at the same place. With every Slicer. Its on every printed object. I almost replaced everything. Even the Motherboard.
Its magic i would say
I hope we find the solution here.
The Lochwürfel otherwise looks good.
What makes me wonder the PA tower seems to look fine.
Did you run it with reduced acceleration?
In your thread on 3DDC we set acceleration and max_accel_to_decel to the same value but still a bit high for a bed slinger.
Did you test it with low acceleration like only 800 mm/s² already or can do so?
I already wanted to suggest turning the object by 45° but you already did it on your own.
This and repositioning the model on the bed should eliminate the rails as root of evil…
The PA tower runs as far as i know with a restriction of 500 mm/s² to increase the material at the corners for better visibility of the right PA value. I will test it with 800 mm/s² and 500 mm/s².
But what makes me wonder is the fact that the indent and so on only happens right before and after a corner and the orbiter 2.0 has a max acceleration of 10000 mm/s². So the extruder should be capable of my current 2100mm/s² with the outline speed of just 40mm/s
There is also almost no ringing with the input shaper values on X and Y and the input shaper results gave me way more acelleration suggestions than 3000 mm/s² on the “slowest” axis. But i use 2100 mm/s².
Even if the 500 mm/s² or 800 will work, i think its just a workaround. Something is not right …
Will post the results with your recommended acceleration soon
Print your model using the exact same slicer and Klipper settings as the PA test object that prints properly. It should look just like the successful PA test object. If so, then start changing your settings one by one until you find the one that causes the corner artifacts.