NEW PART DAY: SILENT STEPPER MOTORS

[Dim3nsioneer 558]

I had the whining also with the DRV8825. A slight adjustment on the current setting made it disappear.

 

[skinnyChicken 1]

has anyone made progress on implementing the watterott parts on the UM2? i was so excited about the noise reduction that i didn't notice instructions were for the UMO and not the UM2 :( my bad. but i still want to reduce the noise level to keep my family from revolt over my printing habit

cheers

[maxy2 0]

 

Yes and no. Yes, x and y axis are important but (no) don't underestimate the extruder motor(s). Just unhook it during a print and you will hear the difference.

 

Just decoupling the feeder from the chassis makes a huge difference. IME, the silentstep stick did not make a huge difference in the sound of retractions, but decoupling the feeder from the chassis did.

 

It was expected that this wouldn't fix the E-Motor noise. Because the E-Motor moves very fast during retraction, Marlin sends two step pulses at a time so the interrupt handler doesn't get called more often than the poor little 8bit AVR can handle. So the output is: "step,step,pause ,step,step,pause, ...". Of course it doesn't help to smooth in-between steps when the steps themselves are irregular. Don't take my word for it, look at the code (stepper.cpp).

[aviphysics 9]

 

 

Yes and no. Yes, x and y axis are important but (no) don't underestimate the extruder motor(s). Just unhook it during a print and you will hear the difference.

 

Just decoupling the feeder from the chassis makes a huge difference. IME, the silentstep stick did not make a huge difference in the sound of retractions, but decoupling the feeder from the chassis did.

 

It was expected that this wouldn't fix the E-Motor noise. Because the E-Motor moves very fast during retraction, Marlin sends two step pulses at a time so the interrupt handler doesn't get called more often than the poor little 8bit AVR can handle. So the output is: "step,step,pause  ,step,step,pause,  ...". Of course it doesn't help to smooth in-between steps when the steps themselves are irregular. Don't take my word for it, look at the code (stepper.cpp).

 

Yuck. Seems like they should have just cut the microsteps to 1/8th and done a single pulse.