Only one stepper is noisy, what settings will quiet it down?

patrickwjoyce

I'm running a Dual Wire Gantry style XY stage, which means that one axis (X) has a 2:1 pulley reduction vs Y. This means that my X motor has twice the steps per mm than Y, so it runs twice as fast to match speeds. I'm running a Duet 2 Maestro with 2.02RC3 firmware, and my Y, Z, and E0 motors are all super quiet (to the point that fans and bearings are my biggest source of noise), but the X motor is extremely loud. I have both X and Y mounted on noise damping mounts, and the noise occurs when the motor is not attached to anything, and when I try different motors on the same axis.

This is the motor that I am using: [PDF warning] http://www.projectsbyzac.com/data/Wantai Stepper Motor 42BYGH610 data.pdf

Here is a video, notice the noise difference on the left to right motion vs the front to back motion; https://www.youtube.com/watch?v=ewG4nRxR5_Y

Here are the motor specific lines from my config.g:

M92 X110.114 Y55.19 Z1513.62 E141.892 ; Set steps per mm, set as if x16 microstepping

M350 X16 Y16 Z32 E32 I1 ; set microstepping

M566 X1800 Y1800 Z100 E100 ; Set maximum instantaneous speed changes (mm/min)

M203 X6000 Y6000 Z600 E2400 ; Set maximum speeds (mm/min)

M201 X400 Y400 Z200 E200 ; Set accelerations (mm/s^2)

M906 X900 Y900 Z900 E900 I30 ; Set motor currents (mA) and motor idle factor in per cent

M84 S15 ; Set idle timeout

Furthermore, here is the output of M569 for each motor:

M569 P3

Drive 3 runs forwards, active low enable, step timing fast, mode stealthChop, ccr 0x00053, toff 3, tblank 0, hstart/hend/hdec 5/0/0, tpwmthrs 2000 (2.6 mm/sec)

M569 P2

Drive 2 runs forwards, active low enable, step timing fast, mode stealthChop, ccr 0x00053, toff 3, tblank 0, hstart/hend/hdec 5/0/0, tpwmthrs 2000 (0.2 mm/sec)

M569 P1

Drive 1 runs forwards, active low enable, step timing fast, mode stealthChop, ccr 0x00053, toff 3, tblank 0, hstart/hend/hdec 5/0/0, tpwmthrs 2000 (6.8 mm/sec)

M569 P0

Drive 0 runs forwards, active low enable, step timing fast, mode stealthChop, ccr 0x00053, toff 3, tblank 0, hstart/hend/hdec 5/0/0, tpwmthrs 2000 (3.4 mm/sec)

Based on other posts, I suspect this is related to the driver switching from StealthChop to SpreadCycle because its much worse at high speeds. I would love to know which settings I should look at to help tune the noise down, ideally without having to reduce speed or axis resolution.

Thanks

dc42

@patrickwjoyce said in Only one stepper is noisy, what settings will quiet it down?:

M569 P0
Drive 0 runs forwards, active low enable, step timing fast, mode stealthChop, ccr 0x00053, toff 3, tblank 0, hstart/hend/hdec 5/0/0, tpwmthrs 2000 (3.4 mm/sec)

You can reduce tpwmthrs using the V parameter in the M569 command, to increase the axis speed at which the driver switches from stealthChop to spreadCycle mode. Watch out for skipped steps at high speeds if you do this.

Reducing Toff to 1 or 2 might also help.

Are you using 24V power?

patrickwjoyce

What would be a good number to aim for for tpwmthrs? I typically run my max speed at 100 mm/s, so it seems like StealthChop is almost always going to be off even for little stuff like homing (5-25 mm/s).

Also, I think I may have separately identified the real problem which is that I'm trying to run a cheap stepper motor at 3300 rpm. The motor is very quiet below 2000 rpm (60 mm/s), and gets quite loud above that number.

I am running 12v power, does that make much of a difference? I'm thinking of getting new steppers, should I consider upgrading to 24v?

genghisnico13

@patrickwjoyce
That motor has a high inductance, switching to 24v will increase the chopping frequency, I'm not sure if 24v is enough to take it above the audible range.
I have motors with 8mH @1khz (datasheets test frequency) inductance and at 12v the chopper frequency is around 10khz. Not sure if the scaling is linear, when I install my 24v supply will see.
I can use them in stealthchop only at high speeds (200mm/s without bearing noises) , you have to reduce the max Acceleration a bit.

dc42

@patrickwjoyce said in Only one stepper is noisy, what settings will quiet it down?:

What would be a good number to aim for for tpwmthrs? I typically run my max speed at 100 mm/s, so it seems like StealthChop is almost always going to be off even for little stuff like homing (5-25 mm/s).

Also, I think I may have separately identified the real problem which is that I'm trying to run a cheap stepper motor at 3300 rpm. The motor is very quiet below 2000 rpm (60 mm/s), and gets quite loud above that number.

I am running 12v power, does that make much of a difference? I'm thinking of getting new steppers, should I consider upgrading to 24v?

If you exceed the speed at which the drivers can provide the current you asked for, then the motor currents become non-sinusoidal and the motor will be noisy. Use the EMF calculator at https://reprapfirmware.org/ to find out the maximum speed before torque starts to drop. Above that speed the motor will be noisy. Increasing VIN to 24V will increase that critical speed.

patrickwjoyce

@dc42 So I just went through the EMF calculator for my motor, but I'm not sure how correct they are since they rely on belt pitch and # of pulley teeth. I use a string drive on a spool so all I have is spool diameter and steps per mm. However, when I manually put my steps per mm at 110 with the rest of my settings, the calculator says that my step pulse freq. is 11 kHz @ 100mm/s, and my torque starts to drop at 65.8 mm/s with 0.8 mA max current provided. This brings up a few questions:

Would changing the microstepping have an effect on the speed at which the motor gets noisy?
Is there a max bandwidth that the motor drivers can handle, and is it related to max current setting in the firmware? I'm wondering if lowering the max current value will let me keep the same speed but quiet down the motor.
Which factor has a bigger effect on the speed at which torque drops: phase inductance or phase resistance? If I do replace the motor I want to make sure I get one that can move at the speeds I desire

dc42

@patrickwjoyce said in Only one stepper is noisy, what settings will quiet it down?:

@dc42 So I just went through the EMF calculator for my motor, but I'm not sure how correct they are since they rely on belt pitch and # of pulley teeth. I use a string drive on a spool so all I have is spool diameter and steps per mm. However, when I manually put my steps per mm at 110 with the rest of my settings, the calculator says that my step pulse freq. is 11 kHz @ 100mm/s, and my torque starts to drop at 65.8 mm/s with 0.8 mA max current provided.

The steps/mm @ x16 microstepping is all the calculator needs to know. The options to enter pulley teeth etc. are for users who are still designing their printers and don't know what the steps/mm will be.

This brings up a few questions:

Would changing the microstepping have an effect on the speed at which the motor gets noisy?

If the noise is due to the driver running out of supply voltage as per the above calculation, then no.

Is there a max bandwidth that the motor drivers can handle, and is it related to max current setting in the firmware? I'm wondering if lowering the max current value will let me keep the same speed but quiet down the motor.

The drivers can handle step frequencies of more than 100kHz.

Lowering the motor current will increase the speed at which the drivers run out of voltage. You will see this if you put a reduced actual current in the calculator.

Which factor has a bigger effect on the speed at which torque drops: phase inductance or phase resistance? If I do replace the motor I want to make sure I get one that can move at the speeds I desire

Inductance.