Z stepperdriver/steppermotor noise

[Henk]

Printer Model

2.4R2

Extruder Type

Clockwork 2

Cooling Type

Stealthburner

Hi all,

Having my fresh build 2.4R2 up and running now I only got a few small issues left one being noisy Z stepperdrivers/steppermotors.
I have a formbot kit, so the setup is a octopus V1.1 with TMC2209 drivers from BTT and moons 1.8 degree steppers.

The issue is that de Z motors are noisy during movement, the X and Y motors with same hardware and config setting however are silent.
Beside that when there is no movement but the motors are locked the (I assume) drivers sound like a fan, when I unlock them the sound is gone.

I have been trying to resolve this issue without result so I am hoping one of you might be able to help me with this.
My settings for the Z steppers in config file is (one of the 4):

## Z1 Stepper - Rear Left
## Connected to MOTOR_3
[stepper_z1]
step_pin: PG4
dir_pin: !PC1
## Octopus 1.0 & 1.1. Octopus PRO 1.0
enable_pin: !PA0
## Octopus PRO 1.1
#enable_pin: !PA2
rotation_distance: 40
gear_ratio: 80:16
microsteps: 32

## Make sure to update below for your relevant driver (2208 or 2209)
[tmc2209 stepper_z1]
uart_pin: PC7
interpolate: false
run_current: 0.8
sense_resistor: 0.110
stealthchop_threshold: 0

What I tried so far is changing the interpolate to true, run_current to 1.4 and stealthchop_threshold to 999999.
I did these changes 1 by 1 and tested after each to see the effect.
Although the sound does change a little from setting to setting it never became quiet like the X and Y are.
And also to note, with the Z belts very lose the loud noise is there too.

Of course I've been out searching on the internet and I found more people with the same problem but none with a solution, I myself have no video/audio of the issue but someone on the klipper forum did post it.
My Z motor sounds exact the same:
Link to the forum topic: https://klipper.discourse.group/t/excessive-motor-noise-vibration-with-tmc2209/12222
Direct link to the video posted by that user: https://www.veed.io/view/d84e75e9-291a-4dfb-9147-81f749d5dff0?panel=share

 

[Henk]

Ok no reponses yet, but I haven't sit still and keep trying to find the issue.
I was wondering if any of you can tell me how hard/easy the gantry should be able to be moved manually? (up and down)
With that I mean fully installed with belts on but the motors not being locked.

 

[Dave32]

Can you also post your [printer] section with the Z velocity and acceleration?

 

[Henk]

Can you also post your [printer] section with the Z velocity and acceleration?

[printer]
kinematics: corexy
max_velocity: 300
max_accel: 3000 #Max 4000
max_z_velocity: 15 #Max 15 for 12V TMC Drivers, can increase for 24V
max_z_accel: 350
square_corner_velocity: 5.0

 

[Ranayna]

This is something i stumbled across while researching Voron printers:

X/Y moves make more sound than they should.... but found a unexpected solution.. (took only 20 hours or so)

My fist build Voron 2.4 is my main printer for printing sets and other Voron 2.4 builds. Yet, what I found out that somehow my newer printers are always WAY more quiet on XY moves with the same config (copy/paste) than my trusty first one... It feels like it needs to work harder on a XY move, doe...

www.teamfdm.com

Maybe worth a try, since you also have inverted the direction of your steppers.

 

[Henk]

This is something i stumbled across while researching Voron printers:

X/Y moves make more sound than they should.... but found a unexpected solution.. (took only 20 hours or so)

My fist build Voron 2.4 is my main printer for printing sets and other Voron 2.4 builds. Yet, what I found out that somehow my newer printers are always WAY more quiet on XY moves with the same config (copy/paste) than my trusty first one... It feels like it needs to work harder on a XY move, doe...

www.teamfdm.com

Maybe worth a try, since you also have inverted the direction of your steppers.

I gave it a try just now and for me it had no effect in sound/vibration difference.
(X and Y are also inverted, didn't change those since those 2 are silent)

 

[NoGuru]

Try Auto tune. https://github.com/andrewmcgr/klipper_tmc_autotune

You can just use it on the steppers you want or all of them. Most of the time it comes down to having the right settings for the motors and the default klipper config just gives you the basics.

 

[Henk]

Try Auto tune. https://github.com/andrewmcgr/klipper_tmc_autotune

You can just use it on the steppers you want or all of them. Most of the time it comes down to having the right settings for the motors and the default klipper config just gives you the basics.

I just gave that a try and the difference is pretty big, vibration is reduced by allot it's not 0 but near that and the sound is totally different, not silent but not a terrible sound anymore.

I also discovered something else, before I tried the auto tune mentioned by NoGuru I was still working on what Dave32 mentioned, the Z velocity and acceleration.
I raised both to max_z_velocity: 50 and max_z_accel: 1000
This gave a very good improvement when I had the gantry move up and down, actually pretty equal to the later auto tune result.
I did notice one thing though, when moving the gantry up and down it of course went faster now and as mention vibration and sound seems much improved, but not during homing.
When I did the homing the Z movement is as loud and vibrating as before, after the homing it's ok when I move it up and down.
So why is that so? has the homing sequence an overwrite on made settings in the config file?
The Auto tune and velocity/accel changes made a big improvement, but the homing behavior makes me feel it's not good yet (I didn't do a QGL yet, it might also do it during that)

 

[NoGuru]

So all motors and have different resonance and they will be different based on power, speed and torque. There is no one size fits all but you can tune to what you can deem acceptable.
You can play with the power settings. On my 2.4 I have TMC 2240 drivers at 24v pushing and my run_current is 1.5 I think with LDO super power motors...on XY.

For your Z you are probably default at run_current .8 I think. Try going up or down a few steps and see what happens. Just remember the faster you go the more current you will need.

 

[Henk]

So all motors and have different resonance and they will be different based on power, speed and torque. There is no one size fits all but you can tune to what you can deem acceptable.
You can play with the power settings. On my 2.4 I have TMC 2240 drivers at 24v pushing and my run_current is 1.5 I think with LDO super power motors...on XY.

For your Z you are probably default at run_current .8 I think. Try going up or down a few steps and see what happens. Just remember the faster you go the more current you will need.

Indeed my z currents are 0.8 at the moment. But all considered with the auto tune settings I now have an acceptable situation and from here I'll try to fine tune to as perfect as can get.

 

[Henk]

I just gave that a try and the difference is pretty big, vibration is reduced by allot it's not 0 but near that and the sound is totally different, not silent but not a terrible sound anymore.

I also discovered something else, before I tried the auto tune mentioned by NoGuru I was still working on what Dave32 mentioned, the Z velocity and acceleration.
I raised both to max_z_velocity: 50 and max_z_accel: 1000
This gave a very good improvement when I had the gantry move up and down, actually pretty equal to the later auto tune result.
I did notice one thing though, when moving the gantry up and down it of course went faster now and as mention vibration and sound seems much improved, but not during homing.
When I did the homing the Z movement is as loud and vibrating as before, after the homing it's ok when I move it up and down.
So why is that so? has the homing sequence an overwrite on made settings in the config file?
The Auto tune and velocity/accel changes made a big improvement, but the homing behavior makes me feel it's not good yet (I didn't do a QGL yet, it might also do it during that)

To answer my own question, I've been reading the code, yes indeed the homing macro does overwrite the values in the config file and restores them after the homing is done, so that explains that behavior.