UMO motion settings

[LePaul]

As I mentioned elsewhere, I've had pretty decent luck setting the acceleration down from 4000 to 3000.

Control->Motion->Accel: 3000
Control->Motion->Vxy-jerk: 12

Not quite sure about Vxy but it was suggested to bring it down from 20 and seems to be working ok!

Any other tweaks in there I should investigate?

[antiklesys]

What improvements have you noticed?
Can you show two comparison prints with the two settings?

[LePaul]

Hi....less ringing effects is what I noticed. I don't think I have the pre-change prints to show the difference tho

[antiklesys]

Well...for the sake of science I'll test this out and add here a picture of the results

[antiklesys]

Tested!

Here's the visual result.
On the left the BEFORE (so original motion settings: Accel: 4000 & Vxy-jerk: 20) on the right the AFTER (Accel: 3000 & Vxy-jerk: 12)

Just me or it's more or less the same?

[Neotko]

To test this you need a fast changing corners print.

Also if the yerk/accel is too low, the changes in flow/speed can make gaps while doing too layers if there's no room for the head to achieve a minimum speed. Also it can make weird effects on corners that affect geometry.

[antiklesys]

Ok, any stl files that you would recommend?
Post a link and I'll test that one.

Just one request: don't make it take longer than 30-40 mins per piece if possible

[Neotko]

I have one that @IRobertI made. I'll link in a few

[Neotko]

https://www.dropbox.com/s/qr1edksfve18u ... t.STL?dl=0

A quick file made by

IRobertI

Some test photos made on the um3 with various test, fixes, experiments and etc etc. Ofc print without top/bottom and 2-3 perimeters. 0.2 layer height somewhere between 0.1-0.2 should show a good debug I think.

Upss sorry I did delete all my photos of ringing tests... Anyhow, is a good test.

At 500 acceleration + 5 jerk the ringing should disappear completely but on the corners you will see a slight deformation of the change of vector, making the square corners a bit rounder than normal (ofc you can't do a perfect 90 degree change).

The part that this test doesn't cover is the top layer small areas with zigzag. The lower the acceleration/jerk the more troubles the feeder will have to make them perfect, leaving gaps (due delay from pressure going to the extruder from the bowden pressure/relax delay). So when the speed goes down too much (too low jerk) artifacts can appear (more visible on top layer).

etc etc etc. Have fun testing!

[antiklesys]

Thanks, will test this out in the next days

[antiklesys]

Ok,

Here's the result with the test model provided from Neotko.
Again on the left is the one with the UMO original settings and on the right the one with the modified settings.

Left--> Accel: 4000 & Vxy-jerk: 20
Right --> Accel: 3000 & Vxy-jerk: 12

To me they still look identical. Am I missing something?

[Neotko]

Umm it should print rotated... I'll search photos of the um3 tests

[gudo]

Hi antiklesys !
Seeing the photos, it seems you have also a Z problem !?
here it's done with the UM2 , the under print shows the improve by lowering the acceleration and jerk settings
I think with UMO you can get something close to that, Neotko would confirm ?

[Neotko]

Ummm I should have explained how to print it... Give me 1h guys =)

[LePaul]

I'm curious why the top of each of his prints look like combing issues?