Hi! Have a weird issue and hope that somebody encountered something similar. When extruding with a Feedrate of 100mm/min everything is OK, so when extruding 100mm it extrudes and retracts exactly 100mm. When using 200mm/min it extrudes 93mm and retracts 97mm. I already grabbed the filament (bowden) to see if it is slipping, but I can not really feel any skipped steps or slipping of the filament. When using 300mm/min it extrudes around 75mm, but retracts 97mm. Using an i3d geared extruder.
inertia
What kind of drive gear are you using? Some drive gears (the knurled drive gears on the ultimakers in particular) are known to deform the filament as it passes over so that the tooth marks end up closer together when there is more back-pressure. All drive gears probably do this to some extent, though some are better than others in this regard.
Make sure your idler is tight enough. If the numbers get better, it’s probably slipping. If they stay near the same, your acceleration might be too high.
@JOHN_YN Not sure of inertia on such low speeds and accelerations
@Whosa_whatsis Drive gear is a custom made. I already thought about machining a new one, but I can not see any irregular marks on the filament …
@Justin_Nesselrotte Idler seems ok. can not see anything slipping. Acceleration is default at 3000mm/s/s. Too hight?
Check the pulse frequenz for extruder by Your board/firmware. A standard NEMA 17 cannot operate with full power over 1000 rpm/3300 pps
@Maxim_Melcher With pps you mean pulses per second, correct? Extruder is a geared one, but apparently I also have 16x microstepping active on that driver. Does microstepping count as pulses? Do you think it could help if I disable microstepping on that driver?
yes, pps are pulses per second. all pulses coming out from Your board counts as pps.
by 500 rpm and 32 microstepping are the pps by 53 kHz. this too much, I think. which steps/mm have You?
1750 per mm. Trying without microstepping in a few minutes now
OMG, 1750 per mm is too much.
Hmm… you are right. My other printer is only using 320 per mm, at roughly the same gear ratio. What is also weird now, I set the drivers to full steps, set the steps to 180 per mm, and extruded 1 mm, which should normally give one full rotation. But it actually gives less then 1/4 …
Did you also tell the firmware to use full steps?
What setting do you mean? I use DRV8825 and the microstepping is set via jumpers on the board directly.
The hardware and software must match in that regard. Not sure what the setting name is. Depends on the firmware.
Marlin. You can check the configuration here if you want. https://github.com/lobermann/Marlin/blob/PrusaI3Hephestos_update/Marlin/Configuration.h. I just found out that the jumpers on the board are not really used, as the pins are connected anyway on the bottom side of the board … cut them open now. Have it running now at 425 steps per mm and going to do some test now. Looks promising so far
Huge improvement! Thanks a lot for the help!
Did a quick test print. Here a picture from it https://photos.google.com/share/AF1QipM46HjUtww9CiYFxPy5k4eoxMu2g4jAuRThh09yE4EwHdCrvCYmFpDAEgE4T1zJVg?key=WlJzU2JOallzeGp3QXk5V0otd2lNdC1DdlJNTFdR . The spikes did not print at all before that, and strings all over the place.