Stepper Vibration/Resonance At Specific Speeds

dc42

@Charlie said in Stepper Vibration/Resonance At Specific Speeds:

The weird thing is, as I increase the stepper current, the vibration moves to a higher gantry speed (for example, increasing the sepper current moves the resonance from 85 mm/s to 100 mm/s.

That sounds like classic stepper motor mid band resonance (also called mid band instability) to me.

Charlie

@dc42 Thank you for your response! I tend to agree. What are my options to try to fix this? My two other printers (Ultimaker UM2 and Creality S5, both with duet boards) do not have this issue at all, though there are a lot of other factors such as Cartesian architecture and stepper type. I can see evidence of the resonance in my prints, and I use this speed range a lot.

Charlie

Any thoughts?

oliof

@Charlie I know this I somewhat out there and most likely cargo culting, but you can try reversing one of the phases (and adjusting the motor definition accordingly). In one community I am dallying around that has improved this kind of issue for a notable amount of people.

Charlie

@oliof Thank you for the input! I will do that and report back.

Charlie

@oliof Unfortunately, reversing the phases and changing the motor direction did not change the noise level. The issue persists.

droftarts

@Charlie As your machine is a CoreXY, I'd check that the belt tension is correct first (too tight can be just as bad as too lose). If you have an accelerometer, and are using a sufficiently new version of RepRapFirmware, you could try using Input Shaping to avoid the resonant frequencies.

It's also worth checking where exactly the ringing is coming from, ie it might be the extruder. See my post here about VFAs: https://forum.duet3d.com/post/321097

Ian

Charlie

@droftarts Thank you for your response! I tensioned the belt using the tool and measurements by PF Makes Tension Meter. I also tried increasing and decreasing the belt tension, with no change to the outcome.

Regarding the Input Shaping (IS), I had input shaping tuned and enabled (M593 P"zvddd" F44.0 S0.1 L0.25). Disabling it did not change the resonance. This makes sense given what @dc42 says here about IS affecting the "acceleration profiles" and (I would assume) not the speed profiles where I am observing the resonance.

Im not completely certain what you mean by

@droftarts said in Stepper Vibration/Resonance At Specific Speeds:

It's also worth checking where exactly the ringing is coming from, ie it might be the extruder.

The resonance is present both during printing, and during traveling at this specific range of speeds, regardless of extruder position or printhead location.

It also happens even if the stepper is removed from the machine, indicating that this is not a belt/machine resonance issue.

engikeneer

@Charlie if you've narrowed it down to the stepper midband resonance (which I'm inclined to agree with), then your options for tuning are limited.

Your stepper is basically acting as a spring damper with a natural frequency that is excited at that speed range. The stiffness comes from the magnetic interaction in the coil, and the mass is the rotational inertia of the stepper. If you increase the stepper current you may effectively make the stepper stiffer and so push the resonance to a higher frequency, else lower current and additing a mass disk to the stepper my drop the resonance to a lower speed/frequency and potentially damp it out a little.

If that doesn't work then either try mounting the steppers on some anti-vibration mounts, or buy some different steppers.

When I built my first machine, the sides of the steppers were butted up against part of the frame, meaning any vibrations were transmitted straight though causing noise. Pulling them out 1mm removed that transfer path and quietened things down a bit

Charlie

@engikeneer That makes a lot of sense. I raised the current to the max rated current of 2,000mA, which gave the best results. Stepping up from 1,000 to 1,900 gave worse and worse resonance until 2,000mA, but the steppers are a tad warm just from a few travel moves. Im not too worried about the motors getting too hot since they are 180C rated, but the plastic Voron mounts may not handle it well on 24hr+ prints.

On the note of changing steppers, what should I be looking for? Ive been looking at the LDO 42STH48-2504AC. I know the low 1.5 mH inductance should be good for high speed, but what specs should I be looking at to reduce the possibility of resonance?

Thank you for the helpful insight!

dc42

My delta printer is noisy when homing and calibrating, which I think is because of md-band resonance. However, it is very quiet when actually printing; so I don't worry about the noise. I guess it may be different for a Cartesian or CoreXY printer in which long travel moves may be noisy too.

oliof

@Charlie the 2504s are a good pick and they like to be driven at spec.