Duet Maestro: Motor noise with default values
-
Okay, so I'm trying a Duet Maestro over the Duet Wifi now. This is a Cartesian build with LDO motors, 24V PSU.
The X and Y motors are the same: 1A rated current, 4.1 Ohm resistance, 47 N-cm holding torque, 11.5mH inductance.
The Z motors I'm missing the inductance on, but they are: 1A rated current, 8 Ohm resistance, and 107.8 N-cm holding torque. It does have Rated Voltage as 8V (not sure if that can be used to figure out the inductance or something else.)
I did see @dc42 link this before: https://reprapfirmware.org/ When I use it for the X and Y I get this:
I have no idea what the "Step pulse frequency" means and why I should want to know about it. I also don't know why the frequencies are listed for the torque drops.
I've set the max current to 600ma and used default settings for M569 which shows:
The motors have a "whine" to them as they move. I've played with setting to stealthChop always on using a command like such: M569 P0 V0. This works great except for moves when doing speeds of 3600mm/min or more, and then it makes a dragging sound. I then experimented with M569 V150 as it says spreadCycle should kick in around 50mm/s
These seems to work, but then I'm back to the annoying as hell high pitch ringy-whiney motor sound.Any help? Thank you!
-
at 3600mm/min and above you start to get resonance sounds from your printer.
first thing you can try is fitting nema motor dampeners.for spreadcycle noise tuning please refer to this article
https://duet3d.dozuki.com/Wiki/Reducing_Stand-Still_Noise -
@veti I tried different TOFF settings for it but they didn't change the noise. The noise isn't there while it's sitting, it's there when it's moving. It's like a high pitched, constant hiss.
I even tried messing with Blanking time and hysteresis, but unfortunately I don't know what I'm doing with those. I was trying to follow along with the Application Note 001 SpreadCycle on Trinamic side too (https://www.trinamic.com/support/app-notes/) but it's a bit confusing with setting hysteresis ...and I don't have an oscilloscope (nor the knowledge on using one )
-
your steppers are high resistance which tend to be noisier in spreadcycle.
Try some of the marlin settings
{ <off_time[1..15]>, <hysteresis_end[-3..12]>, hysteresis_start[1..8] }
#define CHOPPER_DEFAULT_12V { 3, -1, 1 }
#define CHOPPER_DEFAULT_19V { 4, 1, 1 }
#define CHOPPER_DEFAULT_24V { 4, 2, 1 }
#define CHOPPER_DEFAULT_36V { 5, 2, 4 }
#define CHOPPER_PRUSAMK3_24V { 4, 1, 4 }
#define CHOPPER_MARLIN_119 { 5, 2, 3 } -
@veti if I haven't said it yet, thank you for helping me with my posts lol
I didn't see those settings before when I was trying to dig through Prusa's firmware code. There are a ton of settings for the TMC drivers they are using on the MK3, but I don't think we can set those same values, and it looks like those drivers do stealthChop + stallGuard + coolStep, so I doubt they'll work.
My i3 MK2S has been upgraded to an i3 MK2.5 and using someone's external driver board to drive TMC2208s for the X and Y axis ...it runs crazy quiet, but the motors are quite hot. I'm trying to mimic the noise of that one.
I'll give those marlin settings a try. Thanks again! (and I'll make sure to post back so others that search can find this if I come up with a solution.)
-
-
I just realized those values from Marlin are TOFF, HEND, HSTART. I tried them all using:
G91
G1 X50 F4800Still a hiss as it's moving. Hmmm. Guess I need to plug in the other motors and try them (although, they are 0.9deg motors and I have no idea what the inductance is on those either.)
-
Just swapped X motor to this motor (0.9deg instead of 1.8): https://cdn.shopify.com/s/files/1/0887/0138/files/LDO-42STH40-1684MAC-RevA.pdf?10896572297077579660
Using default values that are on the Maestro (step timing fast, mode stealthChop, toff 3, tblank 0, hstart/hend/hdec 5/0/0, tpwmthrs 2000) and trying speeds of F3600, F4800, and F7200 ...no high pitch noise!! And very quiet. Running current of 1176ma (70% of rated current.)
Well, I guess I can chuck these other motors LOL. Shame.
-
you can measure the phase resistance that will give you some indication.
0.9 degree motors in my experience will lead to an improved 3d print.