Voron 0.2 How to get better results for IS and Belt responses

[sage osprey]

Besides more even belt tensioning, is there anything that would be likely to improve results and allow for higher accels/more accurate motion?
v0.2 Formbot kit only ~60 print hours

[sage osprey]

Oh also to note, this was done with panels and top hat not installed so I can retighten screws or whatever I need to do

[wispy peak]

👋

Just saying hi.
Take your time, no rush.

Please ping me or reply to this message so we know u are comfortable that everything from the “How to post here” is provided

[sage osprey]

Yup

well I guess I haven't printed a voron cube with this setup but it's like a step before that so that's fine probably

Also to not all the parts were printed on an ender 3 with a makeshift enclosure and I am looking to reprint most of the parts that carry some mechanical load on this printer

[wispy peak]

Yup as in “im comfortable everything is provided “ or as in “i will let u know once this post is complete”?

[wispy peak]

Not sure I understand.
Can u elaborate that please

[sage osprey]

I mean I want to improve how strongly vibrations are measured before I do another voron cube test print

I'm comfortable everything is provided as far as I can tell what's needed

[opal flint]

@sage osprey can you print and post a voron cube using your current settings please

please post pictures of the four vertical sides

[sage osprey]

okay LOL so I do need to

[opal flint]

Also can you post pictures of your printer please

[sage osprey]

might be a little bit scuffed bc I only did PA properly for ASA and with panels off rn I can only print pla

[opal flint]

PLA is fine.

the general steps we use can be transferred to ASA/ABS as well

[sage osprey]

yeah will just be a bit worse pa value so that will cause some worse print quality but I'll do that give me however long a voron cube takes

[opal flint]

ok

[wispy peak]

The “how to post here” things are requirements and not optional 😉

Thats why i wrote “everything from the…”

[sage osprey]

0 top layers and infill?

cube in 20 minutes

oh man now the cooling isn't enough for PLA at these speeds

[wispy peak]

Exactly as its written in the “how to post here” 😉

[sage osprey]

probably also bc its just so hot right now

[opal flint]

@sage osprey can you zip tie the umbilical to the bowden together near the rear of the printer. Use two zip ties spaced about 1-2 inches apart

[sage osprey]

already one but I can put another further up

[opal flint]

what is your belt tension and how are you measuring it

[sage osprey]

I think it was 115Hz measured with spectroid on my phone

I'll definitely get a better way to measure belt tension

oh and accelerometer mounted by snapping into the screw holes at the front of the toolhead

[opal flint]

do you have a toolhead adxl

[sage osprey]

not permanently mounted no

[opal flint]

can you take a picture of how the adxl is mounted please

[sage osprey]

yeah I have to wait for the print to finish

[opal flint]

ok

[sage osprey]

In order front, left, right, back

Some cooling issues there as well with ASA it's not as much an issue

the cable is very flexible and it's relatively secure like it doesn't wobble around

[opal flint]

I do not think we are going to get good graphs based upon how the addl is mounted, but they are good enough for the print.
You can try to give the addl usb cable more slack. It might cleans things up a little

[sage osprey]

the cable is fine it's very flexible and has enough slack

how should I mount it instead? Reprint different toolhead body with mounting holes to directly screw in the adxl?

[opal flint]

another thing that might be causing the bad graphs, besides the adxl placement and connection, is the umbilical cable might be to short. I am not sure of the exact length it needs to be, but short cables can pull on the toolhead during input shaper test

[sage osprey]

it's just stock formbot kit cable iirc that's also a thing many people replace

[sage osprey]

I'm printing another cube in ASA now bc I have everything calibrated well for that, should be better to look at to isolate any mechanical issues

[wispy peak]

The cube above got no mechanical issues at all, so not sure why a second one should have it

[opal flint]

Sometimes cleaning up pa can show stuff

[sage osprey]

sorry mabye not mechanical issue is the correct word but just VFAs and small vibrations

[wispy peak]

I would be extremely surprised if they got mechanical issues

[wispy peak]

Vfas will only be gone if u print in a green range of the vibration graph in your initial message.
The lower the graph + green the better.

[sage osprey]

outer wall should be in the green range

[wispy peak]

Orca can exclude certain speeds which have a spike in the graph and are in a white range

[wispy peak]

Please verify via slicer preview and compare with your vibration graph

[sage osprey]

Ah actually it's just below the good range

[sage osprey]

probably makes sense to add the 60-105 mm/s as bad speeds in orca then yea

orca only lets me provide a single good range?

[wispy peak]

With run_current, to a certain extent as more current increases heat of motors and drivers, you can change how the graph looks usually

If that doesnt help, tmc autotune allows to change the frequency.
22khz, 20khz or 55khz is what we recommend to start with

[wispy peak]

A single bad range, yea

[sage osprey]

oh

oh overhang speed overrides it?

[wispy peak]

And always verify via your klippy.log that your input_shaper section got 6 config parameters according to your graphs 👍

[sage osprey]

Oh and with the cube here the layer time also lowers the speed

[wispy peak]

Shaper x
Freq x
Damping x

Same for y

[wispy peak]

Overhang, auto cooling, layer time - they all override it

[sage osprey]

oh question, for Y it suggests either MZV or EI, with MZV having some residual vibrations but less smoothing. Is there one option you'd prefer or just personal preference?

[wispy peak]

MZV is most of the time the better shaper/algorithm

[sage osprey]

well maybe getting rid of some of the VFAs would be nice but that's pretty good now lol

they only show up when it's slightly out of focus but that's like really nitpicky tbn no one is noticing that if they aren't looking for it

[opal flint]

What is your current em

[sage osprey]

0.955 I think

[wispy peak]

Please do not “think”, “guess”, “assume” etc.
Verify.
Once verified, double check

We cant work with “i think” properly

[opal flint]

@fierce ingot I think am increase of em might be helpful what do you recommend?
@sage osprey I think increasing em might get rid of some of the Vfa’s

[fierce ingot]

Looks ok to me

To really verify EM, those Ellis' squares and looking at top layer is the way to go. If that shows no gaps, EM is good