Cannot Figure Out Binding

[waxen dock]

Hi,

I am upgrading my V2.4 with mods. It had good IS results before the upgrade, and I am trying to figure out what is causing the worse IS results.

Specs:

V2.4 300 LDO Kit with FYSETC aluminum parts (this was before the upgrade).

-LDO/MasturMynd AWD Kit with 1.8 Kraken Motors
-Stealthchanger LDO Kit

• Anthead with Nitehawk 36 with ADXL 345 onboard. - N3MI-DG Umbilical

Belt tension is measured for value with the PFmakes belt tensioner, and then made to match with the FYSETC belt tension meter.

I have attached photos of the shake tune results and photos of the machine. Please let me know if there's anything for me to do for further information. This is my first time really going through and properly tuning IS. If this is worth anything, the vibrations seem to be in the tool itself, I don't know if there's an easy way to detect failing bearings or something like that...

Thanks in advance!

Machine photos.

[muted token]

Have u tensioner the preload screws

What motors are u running

[vital dome]

@waxen dock FYI jdmontgomery is our Stealthchanger expert, they will support us where our Knowledge might be Limited 👍

[vital dome]

What im seeing aside vibration graphs is that your X backer is installed wrong.
It must always be on the opposite side of the linear rail to counter it

[hazy goblet]

I had basically the same issue. For me it was worn out belts. Around 2700 hrs at that point.

[waxen dock]

the belts are basically brand new with the upgrades since the old ones were too short

[waxen dock]

Kraken 1.8s at 2A

[waxen dock]

thank you, I fixed it!

[waxen dock]

yes, I have

[vital dome]

Welcome.
Are all tests done after running the motor sync addon?

This https://github.com/MRX8024/motors-sync

[waxen dock]

I have not heard of this, let me try it 🙂 and report back.

I currently have synced the belts by following the instructions on the GitHub for the MasturMynd AWD by simply energizing the motors with the front grub screws loose, but this seemed woefully imprecise

[vital dome]

This is pretty essential and must be done before each single print, before each single test and before each single precise move (towards a brush for example)

So always from now on

G28
Motor sync
G28

Then do stuff

This is part of the AWD 101

[waxen dock]

Alright, I synced my motors now, however, I still am getting very similar graphs:

[jovial saffron]

@muted token do you think this has to do with the tap magnets?

[muted token]

It might lots of z lift on y

[waxen dock]

Should I readjust preload?

also, I doubt these things are related, but there is a really high pitched sound coming from the motors when they are not in motion and secondly, i am having issues with the consistency of sensorless, where it will randomly end up homing too early, and sometimes will grind into the end of the axis, despite not changing the value for the SGT. I don't know if these things imply some mechanical issue, but if you all had some thoughts on those issues too, I would appreciate it

I reran the IS after readjusting the preload magnets, and I think these results are better, but still not good?

[vital dome]

do one run with pulse mode please instead of sweeping. see top middle of the graph

here #1308097646786777108 message is how

[waxen dock]

sorry about that, here's the pulsed results

[vital dome]

are u on the latest klipper version?

as well as flashed the latest klipper to your boards?

[waxen dock]

I had to rollback klipper to a version from last week due to the latest update breaking shaketune

host klipper: v0.13.0-403

but my toolboards are slighlty older v.0.12.0-317

I will update that now

[jovial saffron]

do you have toothed idlers to substitute in your xy joints?

can you post your printer.cfg please. I would like to see TMC_autotune and motor config

[waxen dock]

no, but I can get them though

[waxen dock]

here's the printer.cfg, please let me know if you would prefer it in a different format such as in a pastebin or if you want me to just pull out the stuff about the motors

[waxen dock]

I updated my toolboard to v0.13.0-403, will return results soon

[vital dome]

awesome 🙂

[waxen dock]

would the replacements be 20 toothed?

bc if so, the ones I have are slightly too tall, I don't think FYSETC designed for there to be a tootghed idler there... I could replace the XY joints with the chaotic lab or something else if we think that's the problem

[waxen dock]

after the update it looks rather similar...

[vital dome]

yea i agree. Just wanted to make sure u have the li2dw fix from klipper

[waxen dock]

anything I should try next? should I go down the toothed idler path?

[jovial saffron]

Give me 10 and I will make video

[waxen dock]

thank you 🙂

[waxen dock]

Wow that was incredibly informative, thank you so much. The main takeaways: (1) my belts are currently tensioned to 1.9 which is clearly too low for this setup. i have made sure that there is an equal length of belt on both sides. (2) I will make sure both magnets are contacting the screws on the stealtchanger and (3) I will order replacement XY joints (since I never actually had plastic ones) that have proper toothed pulleys in the front. I will also try to print off some plastic ones.

[jovial saffron]

This printer should be able to print plastic xy joints.

the plastic xy joints are for testing

[waxen dock]

I was just going to use my 0.2 though 🙂

I am also slightly less worried about the XY joints than the stealthchanger and belt tension issues since I used these same ones to good results, I admittedly wasn't pushing them as hard as I am now, but they werent problematic before the insatllation of AWD

alright, rerunning input shaper with the higher belt tension resulted in this is this more in line with what you were referencing with there being multiple peaks due to the types of hotend?

y is still a bit messy, but better...

[jovial saffron]

lots of things are not problem in 2wd, but become a problem in awd

@muted token is this looking more like a stealth changer tap issue?

[muted token]

The z lift is what's making me think of the backplate magnets but could be toolboard mount moving

[waxen dock]

I think it might be the tool board mount, there's a lot of wiggle in the mount of the tool board to the orbiter, despite the screws being tightened as there's a plastic bracket. I will try probing with a nozzle mount ADXL345

[waxen dock]

nozzle mounted ADXL345's results, this to me shows that the toolboard mounting is a major problem, although there is still some weirdness with the Y, @jovial saffron do you think this is potentially still the idlers?

reran the other tests