RE: Input Shaper with Klipper Testprint
This looks somehow like a mechanical cause. Belt tension oK? The pully is in the correct position?
I had it once similar but I have not tried different speeds. I have disassembled the Y-axis belt, bearings and pully. And mounted everything according to the instructions. The belt rather a little too tight. Since then it is gone.
RE: Input Shaper with Klipper Testprint
Yes I checked the belt tension and also changed that to tighter and looser. I'll double check the pully, but I also ordered some replacement bearings just in case. Thanks for the tips, I'll let you know what comes out.
RE: Input Shaper with Klipper Testprint
I have 2 GCode for the MK4 high with PLA and PETG.
It is used to measure the resonance.
In this video everything is explained in detail.
Here you can simply enter your measurement and get the frequencies calculated.
Simply post the results here. Would be a contribution to show the effects of input shaper
Thanks for this post this explains a lot. To be honest I had the same view of some of the other people as to what the input shaper was and how it worked. I too thought you needed a sensor on the head for it to work. I think Prusa could have done a better job presenting this feature.
Mini+MK3S+XL 5 Tool
RE: Input Shaper with Klipper Testprint
X = 35.15
Y = 42.73
RE: Input Shaper with Klipper Testprint
How are you sending the gcode commands to the printer and gettting the response? I have been trying to find a way to do that on the mk4 for a while.
RE: Input Shaper with Klipper Testprint
You can add gcode commands to your gcode file as it is generating within PrusaSlicer; Under the Printer Settings Tab; there is a left panel for Custom G-code.
RE:
The test worked great and I decided I want to apply the following shaper values. Unfortunately I don't think the values are applying while printing. I'm on firmware 6.0.0 (release, not alpha/beta)
M593 Y T2 F42.8788
M593 X T2 F41.2121
M500 (tried with and without this)
I also tried with and without T2 in the M593 lines.
I put this in the Start G-Code section in prusaslicer, exported the gcode and did see my change in the file.
The print ends up looking like the default values are still in use. The Tuning->Input Shaping page while printing still show the default values.
Any suggestions?
RE: Input Shaper with Klipper Testprint
Interesting additional point here. Trying it in Octoprint I put the M593 commands in via terminal. I sent:
M593 Y F42.8
and in the terminal I saw
Send: N1033 M593 Y F42.8*18
[...]
Recv: ok
What's the *18 for?
RE: Input Shaper with Klipper Testprint
Hi, try
M593 W instead of M500
buddy-firmware-specific-g-code-commands
"W<1>: Write current input shaper settings to EEPROM."
The test worked great and I decided I want to apply the following shaper values. Unfortunately I don't think the values are applying while printing. I'm on firmware 6.0.0 (release, not alpha/beta)
M593 Y T2 F42.8788
M593 X T2 F41.2121
M500 (tried with and without this)
I also tried with and without T2 in the M593 lines.
I put this in the Start G-Code section in prusaslicer, exported the gcode and did see my change in the file.
The print ends up looking like the default values are still in use. The Tuning->Input Shaping page while printing still show the default values.
Any suggestions?
RE: Input Shaper with Klipper Testprint
I did.
RE: Input Shaper with Klipper Testprint
I chatted Prusa support, they said to email the firmware team. I emailed and didn't receive a response.
M593 doesn't seem to do anything on my prints when configured via PrusaSlicer. But the ringing tower test seemed to work fine. Is there some other command I'm missing?
RE: Input Shaper with Klipper Testprint
New update, chatted support again and they suggested setting the input shaper in each layer. Is that correct?
RE: Input Shaper with Klipper Testprint
The increasing weight of the bed is calculated and adjusted. You could therefore say that the frequency is recalculated with each layer.
For me, I can say that the original values now work amazingly well. You can see from the test print what a wide range is possible over the frequencies.
At the end of the day, it simply works well enough. At least for me.
If I ever have the opportunity to measure with an accellrometer, I will definitely give it a try. The menu in the MK4 suggests that. But since it is still grayed out, I suspect that it does not show the actual values.