Trident 300 Cube

[worldly wave]

Originally was getting better results but realized I had the gantry centered instead of idlers 150mm apart when belt tensioning, so the belts were much tighter. Had 99.7% belt similarity and better shaper results, but seemed wrong to just have tighter belts. Cube was printed using the Y input shaper recommendation from the included image. I didn't use the x recommendation since 3HUMP_EI is not supported.

Not sure what to do to resolve the wonky results.

Printer: VTrident 300cube

Mods: Tap, galvanic support, ptfe tube noodle.

ADXL: adxl345

Belt Tension measurement: Sound Spectrum Analysis app + belt plucking to ~110hz

[narrow cradle]

👋

please reply to this message or ping me, so we know u are ready

[worldly wave]

hey! Thanks for the response, I'm new to this so not sure how this process works

[worldly wave]

whoops, didn't do a reply

[worldly wave]

@narrow cradle was there anything else you needed from me?

[dry vault]

hello

[worldly wave]

hey

[dry vault]

you cube looks okay except for some issues with the retraction in y

your belts and x-axis appear to have a toolhead issue

I would try to solve the overheating/retraction issue first, and if you are happy with it you are done

[worldly wave]

any suggestions on how I would go about resolving that? I was hoping to get a bit more acceleration out of the x-axis

[dry vault]

I would check to see if there are any loose screws

in the toolhead

[worldly wave]

I'll check that out!

actually, what if the thumb screw on the extruder is slightly loose?

[narrow cradle]

u have no real strain relief on your CAN cable. it will break soon or later

install PUG or a PG7 cable gland

[narrow cradle]

recommendation since 3HUMP_EI is not supported
can u elaborate this please?

[worldly wave]

when i added it as an input shaper option it gave me an error saying it isn't valid

[narrow cradle]

dont use machine limit for accel, set them to 20k

use the accel settings in the slicer menu, below speed

[narrow cradle]

thats not possible, 3hump is valid, see https://www.klipper3d.org/Config_Reference.html#input_shaper

[worldly wave]

Unsupported shaper type: 3HUMP_EI is the error i get

[narrow cradle]

pictures are great btw. thanks for actually reading the guidelines

that makes life easier

[narrow cradle]

klipper is case sensitive

[worldly wave]

i mean you're all providing free help, least i can do is be thorough

[narrow cradle]

have a look at the link i posted and copy paste it from there

3hump is not great, we dont want to use it, but its a valid shaper 🙂 thats my point

[worldly wave]

ahh, i tried 3HUMP_EI and 3HUMP_ei

didn't do all lower case 🤦‍♂️

[narrow cradle]

appreciated

[worldly wave]

as is all your help! i really appreciate it

[narrow cradle]

https://www.printables.com/model/1210815-galileo-2-g2e-pug-umbilical-mount-support-to-sb220
to give u an idea. try to find something where the PUG sits lower, on a trident u dont have much space to the top panel.

PUG: https://www.printables.com/model/378567-pug-parametric-umbilical-gland

[worldly wave]

yeah this was the first one i found

although I'm not sure how it'll work with the can cable and the vertical space as you mentioned

[narrow cradle]

on a trident its not that easy, but doable and highly recommended

https://www.printables.com/@Esoterical_306854/collections/2039742

here is a big collection with PUG mounts

this https://www.printables.com/model/747487-clockwork-2-mounting-arm-for-pug-umbilical looks very promising

[worldly wave]

that looks really interesting with looping the cable down and back up, will try that out

[worldly wave]

@narrow cradle may be a dumb question, but just tightening the belts isn't really a solution right? Since it'll just wear them out faster? the results seemed too good when it was over tightened 😅 (old results)

[narrow cradle]

the optimal and max tension is 110hz, yes

u will bend motors shafts if u go higher

[narrow cradle]

also since u are in the US, get yourself the tension meter. its pure gold and a must

https://github.com/Diyshift/3D-Printer/tree/main/GT2 Belt Tension Meter

[worldly wave]

i really want one! but they're all sold out here

id have to order for UK or somewhere else

[narrow cradle]

oh, u can ask in the vendors channels, if not done already, when they restock 🙂

#west3d-🇺🇸-🌐-🌎-🌍 for example

[worldly wave]

yeah i signed up for the west3d email notification for restock

[narrow cradle]

awesome. selfsourcing isnt an option?=

[worldly wave]

theoretically i think i could, i just didn't feel like ordering on amazon. Wanted to try to find the parts from a hardware store but i don't think they have the spring

[narrow cradle]

okay 🙂 its just that in germany we cant order it from amazon 😄 so im always jealous

[worldly wave]

oh yeah, i ran these tests with the doors open since i was printing PLA and it seemed to get too hot.
Updated results with doors closed and tightening stuff and re-tensioning belts:

[worldly wave]

i never even thought of that, yeah they have some of the parts on amazon so maybe i should just take advantage of that 😅

Doors open:

everything seems a bit better, idk why i was hoping to be able to get accels of 10k and 5k for x and y. My kit is an MPX and the documentation mentioned that's what they had but probably depends on environment and stuff

[narrow cradle]

is this a ADXL or li2dw?

the graphs indicate the later because it stops at 190hz

[worldly wave]

I thought it was ADXL, its just whatever is built into the toolhead. My printer.cfg says adxl

[narrow cradle]

which can board do u have exactly?

[worldly wave]

let me double check

[worldly wave]

BTT EBB SB2209 v1.0 with EBB SB0000 CAN v1.0

[narrow cradle]

STM or RP2040 chip?

[worldly wave]

hmmm how do i determine that?

[narrow cradle]

post one of the extruder pins here

if its gpio... its RP2040

if iits PB.. or PA its STM32

[worldly wave]

oh everything is PB

[narrow cradle]

STM32 then 🙂

should be adxl, but weird it stops at 190

[worldly wave]

yeah i'm not sure why. Maybe its a shake&tune setting?

so based on the updated results, is there anything else i can do to improve it?

[worldly wave]

so based on my results would i just take the y acceleration value and use that to replace the 10k values? I'm assuming I just take the lowest value between x and y

[dry vault]

Use ei with ext/int perimeter of 3500

Ei for both x and y

If you sort out the x-axis you likely could get 1000mm/s more

[narrow cradle]

here is the thing

the recommended values only matter for outer walls

because thats where u see ringing.

Lets say u print with >3 walls - that means inner walls can be printer way faster than outer falls because ***if ***inner wall have ringing, the amount of walls will blend that away. does that make sense?

so that being said - infill is the same thing. Do u see it? no. that means u can print infill the faster from all perimeters.
meaning >10k accel at minimum

thats why i said do not use the machine limits because everything else, besides outerwalls, can be printer faster 🙂

and u can not tell your slicer or printer "this is accel for Y and this for X". meaning u have to use the lower value of the two (X & Y) and set that at outer wall accel

Y will always be the slower one - more mass to move

lets say u have 20k in X and 5k in Y.
outer wall accel = 5k

the amount of accel for inner walls is something u have to play with and determine at which point u are not longer happy with how the walls look.
in other words, if we stick to my example values above, 5k for outer walls and 7-8k for inner as a starting point.

[worldly wave]

Thanks for all the info! I’m coming from a sidewinder x2 so these speeds are all new to me

[narrow cradle]

if u print outer walls with 5k but inner with lets say 20k, u will see ringing 🙂

[narrow cradle]

very welcome.
please feel free to ask and ask and ask if something is not 100% clear, u want to confirm u understood it correct or whatever. we dont have any problems with questions but with those who say "i got it" but know they dont and are afraid to ask 🙂

[worldly wave]

Yeah I definitely need to read through all of your last part again

[narrow cradle]

"problem" is the wrong word but i think u got what im saying 😄

no rush, we will be here tomorrow

i asked chatgpt to create a TLDR sort of thing for what i wrote above, maybe that helps reading it with another wording:
🎯 Focus on Outer Walls

• Why Outer Walls Matter: Ringing (wavy patterns caused by vibrations) is most visible on outer walls.
• Solution: Use lower acceleration values for outer walls to reduce ringing.

⚡ Speed Up Everything Else

• Inner Walls & Infill: These parts aren’t visible or get covered up. You can print them much faster.
• Inner Wall Tip: Start with slightly higher acceleration than outer walls (e.g., if outer is 5000 mm/s², try 7000–8000 mm/s² for inner walls).
• Infill Tip: Go even faster here—>10,000 mm/s² is a good starting point.

🧭 Machine Limit vs. Practical Settings

• Don’t just use your printer’s max acceleration limits for everything.
• Only the outer wall needs a conservative setting—others can go faster without affecting print quality.

🧮** X vs. Y Axis Acceleration**

• Klipper/Slicers can’t set different acceleration values for X and Y axes.
• Use the lower of the two for your outer wall setting.
• Example: If X = 20,000 mm/s² and Y = 5,000 mm/s² → set outer wall accel = 5,000.

🔧 Fine-Tuning

• Play with acceleration values for inner walls and infill until you reach a good quality/speed balance.
• There’s no universal perfect setting—adjust based on your printer and desired print quality.

[worldly wave]

Im a person who likes to over optimize so its like im trying to get the most performance but i dont know what realistic limits and expectations are

[narrow cradle]

infill can be printed usually as fast as your hotend can keep up with flowrate 🙂

[worldly wave]

Is there a reason I should use ei over the suggested values? Or is it more precautionary?

[worldly wave]

Oh it wasn’t your wording or anything. I just got busy and haven’t gotten to read it 😅

[narrow cradle]

Sometimes it helps, im not a native English speaker, so GPT can structure and write it better usually, especially when it comes to picking the “right” words

[worldly wave]

It’s all good, you speak at least twice as many languages as me (thanks American education system 🫠)

[dry vault]

Ei is showing 0 residual vibrations (vibr%).I like zero, @ember tide says as long as it is less than5% go ahead and try it.

[worldly wave]

0 does sound very good, I’ll probably play around with it, thanks for the clarification!

[dry vault]

Welcome

[worldly wave]

Probably a broad question but any idea what could be going on with the x? I’m trying to think if it’s worth fine tuning or if the settings I have now are solid enough (they seem like it?)

[dry vault]

Toolhead

Where is the adxl you are using?

[dry vault]

@worldly wave can you take picture of the left and right of your toolhead focusing on the tap magnets please

[worldly wave]

I believe it’s built into the SB2209 board

[worldly wave]

Not sure if these are what you’re looking for

[dry vault]

@worldly wave see above

[worldly wave]

Thank you!

[dry vault]

Welcome

[worldly wave]

The magnets are supposed to touch the screws? I must not have understood that in the manual. Will take apart and test it out

[dry vault]

Don’t need to take apart

Just unscrew the magnet a little bit.

Hear the click

Then tighten up

[worldly wave]

Ahh got that

[worldly wave]

Results afterward, looks like belt frequencies are better, x and y look the same

[dry vault]

It looks like you still have a toolhead issue. Use a flashlight to determine if you magnets are flush with the screws.

[worldly wave]

How can you tell the x has an issue? Assuming because it’s not a single peak?

[worldly wave]

Tweaked them again and maybe I got it this time??

Now y is being weird 🤦🏻‍♂️

[dry vault]

Y looks like a loose screw

[worldly wave]

Huh I only messed with those two and they’re pretty tight, weird

Yeah, no clue what would change when the only variable was adjusting the magnet, unless that’s coming loose during the process

[dry vault]

https://youtu.be/_bvGEa8c0VU

https://youtu.be/IzYNO515-M0

[worldly wave]

Oh that’s a great explanation, thank you for taking the time to go through that. During the last pass to fix the X I did take apart and tighten almost everything in the tool head except the tap, so I’ll take another pass on that and make sure it’s all tight. I think the magnets were in place but it’s hard to completely tell. Again I really appreciate all your help

[dry vault]

Grab a flashlight and shine the light through the magnets on the opposite side you are looking. You do not want to see light between the magnet and screw

[worldly wave]

Yeah magnets seem solid, gonna try to tighten more of the tap stuff

[worldly wave]

@dry vault think this is my last one for tonight. Thank you so much for all the help. It really means a lot. Wish there was some way I could provide some assistance back

[narrow cradle]

those look better

not perfect but better

how about u put those new values into your config and print a new cube, 4.5k outer wall accel

u have to put 6 values for input shaper in total into your config

shaper X
freq X
damping X

shaper Y
freq Y
damping Y

make sure u have all 6 ***before ***printing

[worldly wave]

Should I have turned off the Slow down for better layer cooling?

I had it off for the first print which is probably why there was overheating

[worldly wave]

Before and after:

Also could this gap be causing any of the problems? I’m assuming the bottom gap is causing the top gap after I tighten everything

[worldly wave]

Also realized for this I’m using the 4 magnet Tap R8 setup instead of the 2 magnet 2 screws if that makes any difference

[dry vault]

4 magnets does not really make a difference in regards to resonance

[dry vault]

Your cubes look good.

[worldly wave]

Awesome, thanks again!

[worldly wave]

I haven’t done full calibration for this filament yet but assuming I’m running it too high? 195 seemed fine on the temp tower but maybe need to reduce to 190

[dry vault]

That temp is probably fine.
Might just need to increase fan speed

[worldly wave]

It looks like that was already at fan speed 100%

[narrow cradle]

just to confirm, u had the outer wall accel as i mentioned and increased machine limits to 20k in the slicer?
in the config u added the 6 values from the generated pictures

[worldly wave]

correct

[narrow cradle]

Well done.

The only thing left would be having a look at your vibration profile (shaketune macro) to see if we can improve it.

Otherwise u are good to go🙂

For print tuning we have #slicers_and_print_help

[worldly wave]

Thank you so much for all the help!

[narrow cradle]

As well as ellis guide

[worthy void]

@worldly wave https://ellis3dp.com/Print-Tuning-Guide/

[narrow cradle]

Welcome

Post the Vibration profile once u find the time to 👍

[worldly wave]

also not sure if its good to do or not but it seems like removing the back 2 zip ties smoothed out some of the vibrations?

[worldly wave]

I don't understand any of this 😅

[narrow cradle]

Shaketune have great docs for all macros, have a look at them

Install and configure tmc autotune please

[worthy void]

https://github.com/andrewmcgr/klipper_tmc_autotune

[narrow cradle]

If u need help, please let me know

[worldly wave]

got installed and going through the config.I should switch all my steppers to interpolate: True?

example of my existing stepper

and i would change

[autotune_tmc stepper_x] motor: ldo-42sth48-2004mah

to

[autotune_tmc stepper_x] motor: tmc2209 ?

oh wait..no that doesn't seem right

[worldly wave]

i don't believe my z motors arent specified in the motor_database.cfg.
I believe I have MOONS’ LE174S-T0804-350 but It looks like there's only moons-le174s-t0804-300-ar3-s-150.
Should those be interchangeable with the difference presumably just being the lead screw length?

[worldly wave]

Alright, with the assumption that those motors are interchangeable, I believe I've configured everything

[worldly wave]

Updated vibrations profile:

[dry vault]

I would find the data sheet for the motors

Not sure if it is right

[worldly wave]

for the x/y motors?

or the Z that i was mentioning

[dry vault]

Xy, are they correct

[worldly wave]

Yep

[narrow cradle]

Ignore Z, we only care for X and Y here 🙂

[narrow cradle]

Correct. It must be true

[narrow cradle]

See if 22khz or 20khz looks any better

Add pwm_freq_target to x and y autotune section, with a value of 22e3 and 20e3

22khz —> run vibration profile
Change to 20khz —> run vibration profile

[worldly wave]

22khz and 20khz

[dry vault]

Drop your run current to 0.8

[worldly wave]

assuming that means i need to create a custom motor profile for that?

[dry vault]

No, just want to see what the graph looks like at 0.8 amps

[worldly wave]

how do i change the run current?

[dry vault]

In printer.cfg

Change run_current from 1.2 to 0.8

Just for x and y

[worldly wave]

oh i found it

sorry about that

[dry vault]

Np

[worldly wave]

and run that with 22 and 20khz?

[narrow cradle]

comment the pwm out for now

let it run with 55khz

[worldly wave]

update with 0.8 current:

[dry vault]

This is run this at 23khz please

[dry vault]

The last test has the least amount of motor vibrations

[worldly wave]

So I should stick with that config?

[dry vault]

Yeah

You basically have a dead zone between 55-110 hz for ext perimeters

Did you find your motors in the motor database for x and y motors?

[worldly wave]

Is that normal? Or something I can fix?

[dry vault]

It is motor dependent. Can’t really fix, it is just what it is

Try a cube with autotune on, and the using the same settings for it with autotune off

See which one looks better

[worldly wave]

A full cube or the hollowed out cube?

[dry vault]

Hollowed

Make sure your ext perimeter is not set to the bad speed zone

[worldly wave]

I've had outer wall set to 350 which is probably way too high

[dry vault]

You can run this test from 0 mm/s to 600 mm/s if you like

[worldly wave]

The the vibrations test?

[dry vault]

Yes

[worldly wave]

do I keep the same accels?

[dry vault]

Yes. Do not change the accel of the test

[worldly wave]

Apparently doing 600 with a 2 increment is too much to process afterward for the CB1 😅

[worldly wave]

It appears I should probably stick to lower values

[worldly wave]

Yeah that was my assumption

[worldly wave]

so I'm guessing i should stick to maybe around 200-250 for outer walls. I thought i remember reading to print outer walls faster than inner walls for some reason, Should the inner wall be the same or it doesn't matter as much since the outer wall doesn't appear to have ringing?

[dry vault]

I generally make them the same

For what I print there is no real benefit to go faster

[worldly wave]

Yeah I mean 200 is faster than my old printer does in general, it’s more of the novelty or impatience, but not necessity

Will probably just do 220 for both

[worldly wave]

weird, didn't really notice a difference between changing the inner down to 220 from 400.
also didn't notice a difference between autotune and no autotune. Not sure if i should disable it or not but may live it for now with 220 inner/outer since it seems to work fine

[worldly wave]

I added on a strain relief for the CAN cable but it seems like that made everything the input shaper results worse again.
Is that just something that’s worth the tradeoff?

[worldly wave]

@dry vault sorry to bother you. Would you have any insight on this? Assuming it’s just the con of adding an additional piece to the tool head.

[dry vault]

If you removed the 90 degree tube from the top of the toolhead what does it look like

[dry vault]

Try print a cube with extra perimeter of about 3500

That peak is probably tap

[worldly wave]

Using the updated input shaper and 4 perimeters?

[worldly wave]

Some wonky results but assuming it’s due more to using ABS and it maybe getting cooled too much or something

[worldly wave]

Oh that edging is PA? I thought I tuned it right

[worldly wave]

Yeah you’re right, I must have had my z offset too low when I first tuned it. Better but still not perfect

[narrow cradle]

Looks wonderful

No ringing

How fast did u print outer walls in accel?

[worldly wave]

outer and inner walls at 220mm/s at 3500 accel

[worldly wave]

And assuming you mean wondering in terms of the ringing? Seems like I still need some tweaking for the rest of it but it’s great that it seems like the resonance is okay

[narrow cradle]

And assuming you mean wondering in terms of the ringing?
yes correct, i dont care about the rest 🙂 as this channel is for resonance 🙂 for print tuning we have #slicers_and_print_help

@dry vault do u think we can do more here or are we at the point where its "enjoy your printer" ?

[worldly wave]

I'm probably going to have to redo some of it because i didn't print the right toolhead to do a filamatrix, but i think with your suggestions i'll be able to test and tweak it as needed

[dry vault]

I think now it is just regular printer tuning. Which is half the joy of printing

[narrow cradle]

ist been 17 days, how are things going? any update for us? 🙂

[worldly wave]

heeyyy! Things have been going well! Were going well, i had a boxturtle kit i was setting up and had to redo the whole toolhead for a filamatrix and i fried my sb2209 so trying to fix that now. In terms of printing before that everything was solid but I'm going to have to go through the whole tuning process again since I made all these changes.

Oddly noticed that the toolhead was shifting a little after replacing the extruder and i guess the TAP rail got loose, not sure how that happens or how often I should need to tighten that.

tldr: your help made everything work great, i keep doing new things and breaking stuff 😂

[narrow cradle]

awesome, glad to hear. so we can mark this post as SOLVED? 🙂

[worldly wave]

yep!

[worldly wave]

@narrow cradle If I had another question about input shaper results since adding a filamatrix, do I need to create a new post or could I just ask here?

[narrow cradle]

Depends on how bad the graphs look
Im kidding, go ahead

[worldly wave]

thanks! just wasn't sure about this second peak in the y

[narrow cradle]

U did only install filamatrix, Everything else is as it was?

No new xy joints, bearings, etc

[worldly wave]

well to do the filamatrix I had to do a new clockwork and hotend holder for the dragon hf, so most of the toolhead is kinda new now. Tap and the stealthburner cover stayed the same

[dry vault]

is there filament in the extruder?

[worldly wave]

there was not, I'll test it with filament

[worldly wave]

Update with filament in extruder, seems pretty same-y. Is this something I need to fix or will it operate fine with the shaper settings?

[dry vault]

that second peak is likely what is causing your low accels in y.
However, lets print a cube using EI for both x and y setting int/ext perimeter accel to 3.5k

voron cube = 3 bottom/perimeter, 0 top/infill

[worldly wave]

Think it seems mostly good?

[dry vault]

Yup. Just some fine tuning but looks good

[worldly wave]

Thank you! Apparently the Y shaper settings are almost identical to what I had before, guess I should've double checked before taking all your time again 😅