V2.4 AWD Shaketune Results

[gritty snow]

Printer: Voron V2.4
Gantry: Monolith Gantry 9mm AWD
XY Motors: 48 V
Carbon Drive at +-230Hz

[resonance_tester]
accel_chip: beacon
probe_points: 175,175,20
sweeping_period: 0

Note / Observations:
The build was recently rebuilt, and I’m still learning how to fully interpret all results.

What I currently notice:

In the Input Shaper measurements, the Y-axis shows higher values than X.
I’m unsure whether this is expected behavior or indicates an underlying issue.

In the Vibration Profile, the motor currents differ slightly.
Does this point to a potential problem or is this considered normal?

On the test cube, there is clear ringing/ghosting on the Y-axis.
Changes in print speed seem to have little effect.

Overall, the printer feels like it produces too many / too strong vibrations.

I’d appreciate any feedback on where I should focus to further improve the system.

[turbid falcon]

👋

A few things u have to change:

• use toolhead adxl for all tests
• ignore axies map, its useless currently

[gritty snow]

Hi @turbid falcon
thank you for your support!

I’ve already tried multiple times to use the accelerometer on the toolhead board.
The installed board is an EBB36 Gen2.

In the example configuration from GitHub, it looks like this:

[lis2dw]
cs_pin: EBB:PB1
spi_bus: spi2_PB2_PB11_PB10
axes_map: x,y,z

When I adjust my printer.cfg to the following setup:

##--------------------------------------------------------------------
[mcu]
serial: /dev/serial/by-id/usb-Klipper_stm32h723xx_230031000A51313339373836-if00
restart_method: command
##--------------------------------------------------------------------

[beacon]
serial: /dev/serial/by-id/usb-Beacon_Beacon_RevH_459D17E25154354D38202020FF0A3D25-if00
x_offset: 0
y_offset: 30
mesh_main_direction: x
mesh_runs: 2
speed: 10.0
#accel_axes_map: x, y, z

#[resonance_tester]
#accel_chip: beacon
#probe_points: 175, 175, 20
#sweeping_period: 0

[mcu EBB]
serial: /dev/serial/by-id/usb-Klipper_stm32g0b1xx_0B001C000450315741363420-if00
#canbus_uuid: a62a88bb28e3

[resonance_tester]
accel_chip: lis2dw
probe_points: 175, 175, 20
sweeping_period: 0

[lis2dw]
cs_pin: EBB:PB1
spi_bus: spi2_PB2_PB11_PB10
axes_map: x,y,z
##----

I get the following error:
mux command ACCELEROMETER_MEASURE CHIP None already registered ({'lis2dw': <bound method AccelCommandHelper.cmd_ACCELEROMETER_MEASURE of <klippy.extras.adxl345.AccelCommandHelper object at 0x7f5058f8d0>>, None: <bound method AccelCommandHelper.cmd_ACCELEROMETER_MEASURE of <klippy.extras.adxl345.AccelCommandHelper object at 0x7f5058f8d0>>, 'beacon': <bound method AccelCommandHelper.cmd_ACCELEROMETER_MEASURE of <klippy.extras.adxl345.AccelCommandHelper object at 0x7f68e47150>>}) Once the underlying issue is corrected, use the "RESTART" command to reload the config and restart the host software. Printer is halted

I’m not able to understand what the root cause is here.
Could you help me identify what I’m doing wrong?

Thanks in advance!

[lilac stirrup]

@gritty snow When posting configs or logs, please surround with code fences (```) so that Discord formats them correctly. Example:
```ini
[mcu]
serial: /dev/serial/by-id/usb-klipper-12345-if00
```

[mcu]
serial: /dev/serial/by-id/usb-klipper-12345-if00

[turbid falcon]

never seen this one

try removing beacons adxl section

oh u did, nvm

[dusty wave]

Your belts are off. Did you cut them the same length, and have the same number of teeth on their respective side

Your spi bus on lis2dw does not look right

[turbid falcon]

apparently it is https://github.com/bigtreetech/EBB/blob/eb85a78b40b8020fedb2a138017701616c2c95b6/EBB_GEN2/EBB36_GEN2/sample-bigtreetech-ebb36-gen2-v1.0.cfg#L19

[gritty snow]

Thank you for your messages.

The lis2dw configuration was taken directly from the BTT GitHub:

[lis2dw]
cs_pin: EBB:PB1
spi_bus: spi2_PB2_PB11_PB10
axes_map: x,y,z

What exactly do I need to change or verify here?
I’m not very familiar with SPI bus / pin mapping, unfortunately.

Regarding the belts:
They were originally cut to the same length.
After checking again, I’m no longer completely sure — it’s possible that one belt is off by one tooth.

Thanks for any guidance!

[turbid falcon]

nice ai prompt 😉

[gritty snow]

hehe, ich brauche Hilfe in Englisch..

[turbid falcon]

Das einzige was in am würfel sehe sind die vertikalen linien, das sieht für mich aber nicht nach ringing aus. eher PA, layer time oder so was.

english:
the only thing i see on the cube are those vertical lines, doesnt look like ringing to me

[turbid falcon]

alles gut 🙂 war nur lustig mit dem Ki gedöns davor noch

unter berücksichtigung des Channel namens, #1308016330326413393 , kümmern wir uns hier um Basic hilfe. Die Graphen selbst sehen gut aus, die werte ebenso, der würfel hat kein ringing. Dementsprechend sehe ich kein bedarf unserer hilfe hier.
Beantrage eine Serial, dann öffnet sich das Tor um da abseits von "Basic" nachzuforschen 😉 ist nur ein kurzes video und zettel...

[dusty wave]

speed

potentially EM

[turbid falcon]

yea possible, the vibration profile speed range thingy is 1e9, so quite noisy.

[dusty wave]

I would work on getting those belts even in "compare belts"

[turbid falcon]

Would you say thats worth it, considering our basic help here? the cube looks good imo

[dusty wave]

It is up to the user, but in regards to basic help, getting belts looking equal I would say is basic help

the belts could be the potential cause in the angular speed energy sub-graph (top right) of vibrations profile

[gritty snow]

Thanks for all the input, I’ll go through these points carefully again.
I should mention that many settings were carried over from my stock gantry without a full recalibration.

I’ve also realized that I’m no longer very interested in pushing extreme speeds, but rather in achieving smooth and clean print quality.

Would you recommend going back to the stock aluminum extrusion gantry in this case?

[dusty wave]

you should be able to get clean prints with your gantry

awd should be able to do 10k perimters, but don't expect crazy speeds

[gritty snow]

After some experimentation, I have arrived at these results. Since the last test, the printer has completed around 60 hours of printing. I am satisfied with the print quality and will keep the setup as it is for now.

The belts are correctly matched in length at the tooth level. I cannot explain why there are two different curves here.

One thing still puzzles me: the Input Shaper graphs may be swapped. It looks like X is showing Y and vice versa. Am I correct in this assumption?

[dusty wave]

Your belts are off likely cause of belt tension.
Why do you think your input shaper axis are swapped

[gritty snow]

As far as I understand it, I see higher values for Y instead of X. That doesn’t seem to make sense to me.

[turbid falcon]

Could we get a fresh picture from the front of the printer please?

[dusty wave]

Look at ei or mzv not zv

[turbid falcon]

do u have galvanics arm installed?

https://www.printables.com/model/978123-voron-bowden-ptfe-tube-guide-arm-and-canbus-cable

[gritty snow]

will replace the cable routing again and install it together with the new arm.

[gritty snow]

I disassembled and rebuilt the gantry once more. During this process, I noticed that in several places washers between the bearings were missing, which caused the bearings to rub against each other.

After rebuilding the gantry, I ran Shaketune again — the result shown above is from the first run.

I expect that many of the issues I was experiencing will be resolved now that the gantry has been rebuilt correctly.

Thank you for your support.

[dusty wave]

Looks good