Z-axis noisy  

  RSS
carl
 carl
(@carl)
New Member

I just finished assembly of the MK3S kit. Everything is working but compared to the X-axis and Y-axis the Z-axis is very loud. 

I already did these checks, put the problem is still there.

- removed the trapezoidal nuts 

- tested if the slide without resistance on the z-axis, this is OK no friction, added high temperature synthetic oil lubricant

- tested if the rotate free on the Z-axis lead screw, this is OK no friction, added high temperature synthetic oil lubricant

- checked the Z-axis motor mounts and Z-axis top parts

I think there is some misalignment between the Z-axis lead screws and the Z-axis rods but I do not see any way to tune these. 

Any suggestions?

Carl 

Posted : 19/06/2019 11:28 am
jack
 jack
(@jack)
Trusted Member

carl... Perhaps this is a different issue, but here's a problem I had with z Axis noise and how I solved it: https://forum.prusaprinters.org/forum/original-prusa-i3-mk3s-mk3-assembly-and-first-prints-troubleshooting/z-axis-noise-problem-solved/ .  The z Axis steppers do not seem to run as smoothly as the x and y steppers on my machine. Maybe it just the fact that there are two of them stepping simultaneously. Hope this helps...

...there are only 10 kinds of people in this world--those who know binary and those who don't......
Posted : 19/06/2019 12:45 pm
carl
 carl
(@carl)
New Member

So you put dampers between the motors, the Z-axis top and the aluminum frame?

 

Posted : 19/06/2019 12:51 pm
carl
 carl
(@carl)
New Member

I found the problem. Unfortunately the a problem in the EINSY miniRAMBo board.

So, what is causing the problem. There are two Z stepper motors on the same TMC2130-LA driver. The result is uneven stepping between the left and right Z-stepper motor. Because of the back EMF (electro magnetic force) the stepping will never be smooth. Using a TMC2130-LA is probably a bad idea. 

This post was modified 1 year ago by carl
Posted : 19/06/2019 2:16 pm
Tim
(@tim-m30)
Illustrious Member
Posted by: carl

I found the problem. Unfortunately the a problem in the EINSY miniRAMBo board.

So, what is causing the problem. There are two Z stepper motors on the same TMC2130-LA driver. The result is uneven stepping between the left and right Z-stepper motor. Because of the back EMF (electro magnetic force) the stepping will never be smooth. Using a TMC2130-LA is probably a bad idea. 

Except that a lot of us are using that board and NOT having noise issues.   I suspect there is something wrong in your build.  For example, the Delrin T-nuts are self lubricating, no oil needed.  Bearings specify grade 2 grease.  And a common problem with the Z-Axis is the dust covers are too close to the motors and cause binding on one side.  Too much X-Axis belt tension can also cause Z-Axis issues.  Another less likely possibility is a broken wire on one of the motors.  

 

This post was modified 1 year ago by Tim
It is always wise to get more than one opinion......
Posted : 19/06/2019 7:29 pm
jack
 jack
(@jack)
Trusted Member

So you put dampers between the motors, the Z-axis top and the aluminum frame?

carl... No, the rattles were actually inside the LCD module due to loose parts. That's where I put the felt. The rattles were excited by the z Axis vibrations and added quite a bit of noise. You may have pinpointed why the z Axis motors seem to vibrate more than any other motor, however. I use steppers in my Arduino projects quite a bit, and they can be a mystery to get running smoothly. I suppose it was easier running two motors rather than just one with a belt on top. Maybe there was too much play in that solution, or side-to-side adjustment difficulties. Could be an easy mechanical mod, but I ain't gonna do it!

...there are only 10 kinds of people in this world--those who know binary and those who don't......
Posted : 19/06/2019 8:51 pm
Tim
(@tim-m30)
Illustrious Member

A belt drive Z is problematic on several levels. #1, belt stretch causes a X-Axis misalignment that can't easily be tweaked out. Imagine having to manually "recalibrate" Z lead screws to do an X-Axis level every other print. Arggh.  I'd stop printing.  And, there really isn't anything wrong with parallel drive for stepper motors. Sure, if one is carrying a higher load than the other, problems can crop up. But in this application there isn't any reason for load imbalance except an assembly issue. 

It is always wise to get more than one opinion......
Posted : 19/06/2019 9:04 pm
carl
 carl
(@carl)
New Member

I think I found the issue. I measured the coil resistance of all my steppers. 

The combination of one stepper driver and one of the steppers that is probably outside the specs 

stepper

A coil

 

B coil

 

 

X-axis

6,4

Ω

6,4

Ω

100%

Y-axis

6,4

Ω

6,4

Ω

100%

Z-axis right

8,2

Ω

7,8

Ω

105%

Z-axis left

7,9

Ω

7,9

Ω

100%

Extruder

6,5

Ω

6,5

Ω

100%

Posted : 21/06/2019 8:45 am
Tim
(@tim-m30)
Illustrious Member

Are you using 4 wire measurements? If not, that 0.3 ohm is meaningless.

If using relative measurements, if may describe a crimp issue... but I still doubt the numbers unless great care was taken to minimize errors.

This post was modified 1 year ago by Tim
It is always wise to get more than one opinion......
Posted : 21/06/2019 6:41 pm
jack
 jack
(@jack)
Trusted Member

Z-axis is very loud

carl...I found that one very powerful way to diminish z Axis noise is to modify a certain value in the PrusaSlicer program before slicing. Go to Printer Settings>>Machine Limits>>Maximum Feed Rate Z, and change the default value (12) to something like 6. Slice the model and print it, and you should have virtually no z Axis noise. (the z Axis motion will be slow, of course). The problem comes from the stepper motor exciting some natural structural frequencies in the printer. I used an acoustic spectrum analyzer to view what was happening, and there are many natural frequencies in the 1-2 kHz region that get excited by vibrations of the steppers if the speed is set to 12. Lowering that speed lowers the excitation frequency, and probably its magnitude, so the natural modes are not excited. I played with this variable, and found that a value of 10 is still quiet, with just little loss of speed. Give it a try...

...there are only 10 kinds of people in this world--those who know binary and those who don't......
Posted : 30/06/2019 3:02 pm
Share:

Please Login or Register