which Input shaping

[Firefly]

which to pick ? reduction factor closer to 0 is what I am to pick?

ZVD good enough or go with ZVDD?

[dc42]

@Firefly ignore the peaks at 80Hz and above, they are probably vibration from the fan. The main peak you should target is the Y ringing at around 28Hz. I suggest you try ZVD or MZV centred at 28Hz.

If after that you find that the Y ringing is cancelled but you also need to target the small X ringing at 40Hz, then try ZVDD or EI2 centred at about 32Hz.

btw the default damping factor of 0.1 is too high in many cases. You may get better results using 0.0.

[Firefly]

@dc42 Thank you for the reply, I have no clue what I am doing and did not realize I was suppose to change the Hz. so is this what should be done?

[dc42]

@Firefly yes, that looks about right.

[Reine]

Sorry for the hijack but I've been struggling with ringing for four days now and printed more ringing towers than a person should ever have todo

This is how my RatRig VCore 3.1 400 printer looks with no shaping applied. It varies a bit but in general.

I'm not sure I have the reference right but on pure Y moves, that face have not much ringing at all and with shaping the little that was there is mostly gone.
But straight X moves have a lot of ringing and no matter what I do I can't cancel it out.

Given that the graph shows a large spike on Y and that Klipper ringing tower marks the side with ringing Y I guess I somehow have it mixed up, but that's the side where the print head only moves in X direction. Maybe it's some CoreXY thing, idk.. How ever..

How should I set up IS based on that graph?

My last attempt was with ZVD at 58Hz and that looked like this.
( PA 0.045 and Jerk at 150 )

(Klipper Y actual X only move)

(Klipper X actual Y only move)

[Reine]

Based on @dc42 suggestions above I made some own assumptions and tests.

First my original ZVD @ 58Hz

ZVD @ 47Hz

ZVDD @ 47Hz

EI2 @ 47Hz

Should add I'm on v3.4.6 with 6HZ + 1LC + PanelDue 7i + SBC.

[dc42]

@Reine it's common for vibrations on the X axis to cause ringing artefacts on the U face, and vice versa.

We are doing some more work on IS in RRF 3.5 to give better results when there are short moves in one direction followed by long moves at 90degrees to that direction. The ringing tower prints deliberately include these shapes.

[Reine]

@dc42 I'm looking forward to trying that. Any rough eta on 3.5?

[dc42]

@Reine we hope to release 3.5.0-rc1 next week.

[Firefly]

@dc42 when I when to use the printer again, I saw the message "Input Shaping is not configured. Record a new Motion Profile to set it up" seems I lost it, it happened before and I thought I deleted it by accident, but not this time. printer was off for awhile and don't see how to recall the last motion profile. I already removed the input shaping sensor, but if I do this again what is the proper way to save and make it stay?