Voron 2.4 belt tension inconsistence detected by klippain-shaketune

[quasi carbon]

Hello, I try to figure out why my belt inconsistence when tested by klippain-shaketune's marcos COMPARE_BELTS_RESPONSES ?

My machine have already changed new belts, new gates idler, bfi, new motors.

• voron 2.4 300
• adxl 345
• x, y motor ldo-42sth48-2504ac with tmc2240 drivers by klipper_tmc_autotune
• BFI, klicky-probe with pcb-probe
• x,y belts tension are both 130 hz , measuring by gates carbon drive app and GT2 belt tension meter
• idlers tooth cleaned
• belts path smooth

[mcu rpi]
serial: /tmp/klipper_host_mcu

[adxl345]
cs_pin: rpi:None

[resonance_tester]
accel_chip: adxl345
probe_points:
    148, 146, 20  # an example
accel_per_hz: 100
sweeping_accel: 400
sweeping_period: 0

[slender epoch]

Can you post graphs and pictures of the printer please

[quasi carbon]

posting

[quasi carbon]

At first, because of the VFA, I dismantled the gantry and replaced the belts、gates idlers、bfi、 motor.s So now I use Shaketune to test the belt but haven't printed anything yet. Sorry, there is no cube effect.

[quasi carbon]

can you see it on this pic?

[slender epoch]

Can you post input shaper results please

[quasi carbon]

wait for running.

[slender epoch]

Ping me when you get the results. I get lost in discord

[eternal lark]

before i join this i need to know, ELRS yes or no?

[slender epoch]

@eternal lark Elrs?

[quasi carbon]

ELRS ?

short for?

[eternal lark]

ExpressLRS https://www.expresslrs.org/

[quasi carbon]

no, my quad do not use it.

I use black sheep 915.

[slender epoch]

@eternal lark what is this?

[eternal lark]

https://tenor.com/view/veeunus-spongebob-spongebob-squarepants-squidward-done-gif-24674884

[eternal lark]

its a radio protocol mainly for FPV drones 😄

[quasi carbon]

Do you like my avatar?

@slender epoch

[slender epoch]

do you have a fan on when you are testing

[quasi carbon]

stepper fan is running

https://www.printables.com/model/551802-fysetc-stepper-driver-cooling-fans

[slender epoch]

turn off/unplug all fans and run the test again. Want to see if the vertical spike in the spectrogram will go away

[quasi carbon]

how about pi fan?

will it overheat?

@slender epoch I think a motor idler is just ok, photo angle problem

[slender epoch]

okay

[slender epoch]

it will be fine for the test

are those the only fans going, or is there a toolhead fan going

[quasi carbon]

no toolhead fan.

only pi fan, controller fan, stepper fan.

just now I close stepper fan, the x, y driver shows temperature 100.

@slender epoch is the temperature safe?

@slender epoch without any fans

[slender epoch]

Turning off the fans looked like it calmed it down a bit

I think everything we are seeing is due to the adxl cable

You can try a different adxl that uses a usb cable, or Set input shaper to ei for both x and y and print a voron cube, limiting ext/int accel to 4k.
Voron cube= 3 bottom/perimeter, 0 top/infill

[quasi carbon]

When I zipped this, belts compare looks better

[quasi carbon]

I don't known how to fix it?

[slender epoch]

Do you have another adxl?

[slender epoch]

@quasi carbon did you see this

[quasi carbon]

I have another same adxl, not usb one.

[quasi carbon]

Ok, can I print it pla? Print abs I need close my panels.

[slender epoch]

Yes you can print the cube with pla

We just want to get a gas.ine print to see where we are at

[quasi carbon]

@slender epoch 0 infill make it unsupport?

[slender epoch]

@quasi carbon It will print

[quasi carbon]

@slender epoch My pla filament has a bit wet, do I need dry it and print a new one?

[slender epoch]

You do not have ny resonance

For the other issues in the print I would head to #slicers_and_print_help

[quasi carbon]

can you see the vfa?

[slender epoch]

No Vfa’s.

[quasi carbon]

how about this?

[slender epoch]

Same cause as your other box.

[quasi carbon]

pa? em?

[slender epoch]

Pressure advance
Extrusion multiplier aka flow

[quasi carbon]

My input shap config is

[input_shaper]
shaper_freq_x: 58  # frequency for the X mark of the test model
shaper_type_x = ei
shaper_freq_y: 50.2  # frequency for the Y mark of the test model
shaper_type_y = ei

I want to print 200 speed

acc 5000

[slender epoch]

You can include damping ratio

[quasi carbon]

Is it fine to print that speed/

[slender epoch]

200 with pla will be tough without more cooling

[quasi carbon]

print abs

[slender epoch]

Abs should be able to do it, will need cooling

At 5k accel I would print a cube at that speed to see if artifacts emerge

[quasi carbon]

I want to know as my input sharp test, why don't you choose mzv?

[slender epoch]

You have a lot of junk in your graphs, and I know ei will produce better results than mzv with minimal downside

[quasi carbon]

But ev limit the accel

[quasi carbon]

Where can I learn this info? blog?

[slender epoch]

https://www.youtube.com/@The_original_Reth

[quasi carbon]

Oh, you are Reth?

This is your channel.

[slender epoch]

Yes

[quasi carbon]

Thank you for helping me try out and share this lesson.

[slender epoch]

Welcome

[quasi carbon]

Next I will close the panel and use the correct PA and EM of ABS. If I have any other questions, can I continue to ask here?

[eternal lark]

If its about resonance, yes 🙂

[quasi carbon]

Print speed 200mm/s, accel is 4000, y axis shows vfa.

[slender epoch]

Run vibrations profile please

[quasi carbon]

Wait me read the document.

[quasi carbon]

CREATE_VIBRATIONS_PROFILE SIZE=100 Z_HEIGHT=20 MAX_SPEED=200 SPEED_INCREMENT=2 ACCEL=4000 TRAVEL_SPEED=450

@slender epoch

[slender epoch]

Redo this at the default accel.
Everything needs to be default

The goal of this test is to measure the noise generated by the motors.
However, raising the accel turns it more into an input shaper test.
I believe the default accel is 3000

[slender epoch]

you can just hit the button without doing the pull down menu

[slender epoch]

for organic prints (things with lots of curves and stuff print in the green zone in the global speed energy profile
For structural parts print between the peaks (in the valleys) of the angular speed energy profile.
---for the angular speed energy profile use the graph which corresponds to the orientation of print

[quasi carbon]

Let me take in what I learned today, I don’t even know where I am. Sorry.

[quasi carbon]

I have read documentation many times. What's the organic prints and structural parts prints? I think I should set my silcer speed in the green zone. So speed 200mm/s is ok? Why the vfa comes out?

Does this graph tell me what is wrong with my machine and how to improve it? I learned from the documentation that I should use the appropriate speed?

[slender epoch]

organic prints are things with lots of curves. while structure prints have straight sides

think toys as organic, and printer parts as structural. Although there is some overlap

with structural parts you have more speed range available than you do with organic. This is because one is not so concerned with all the angles the toolhead has to make

for example if I print cycliner I have to wory about all the angles the toolhead has to make and choose from a limited number of speeds becasue each angle of the toolhead will have different vibrations, and the speed I choose hopefully will not cause VFA

whereas in a structural part, think box, one is generally moving the toolhead in x and y only, with slight 45 to connect the x and y, such that I have more freedom to choose a speed.

this is because I do not have to worry about all the other angles

see polar angle energy profile and polar vibrations heatmap

[quasi carbon]

Polar angle energy profile shows my belts is 100 percent symmetry, polar vibrations heatmap I don't know how to read.

[quasi carbon]

So when I print something with a complex structure I should pay attention to the speed, but simple straight lines like box don't need to pay attention?

[slender epoch]

Almost

With complex shapes (organic) speed is very important to make sure Vfa’s are not present.
With simple (box like, structural) shapes speed is not as important, as you have more ranges of speeds, but it is still important

[quasi carbon]

Green zones contain speed 200mm/s, how can I print 200mm/s?

[slender epoch]

Sometimes you can’t. Just because your printer can reach the speed/accel does not mean you can print at that speed

[quasi carbon]

How can I improve my printer to reach that speed? I have seen many printers running more faster than mine.

[slender epoch]

What is your goal? Good prints, faster printing, or time/money pick two

Run vibrations up to 400mm/s

Pla is limited, abs you can probably go faster

[quasi carbon]

Print abs 200mm/s with good outfit.

[quasi carbon]

increment 2 or 10?

[quasi carbon]

// Machine estimated vibration symmetry: 100.0%
16:34:19 
// Vibrations peaks detected: 4 @ 58.1, 79.3, 296.1, 307.1 mm/s (avoid setting a speed near these values in your slicer print profile)
16:34:19 
// Lowest vibrations speeds (4 ranges sorted from best to worse):
16:34:19 
// 1: 2.0 to 54.8 mm/s
16:34:19 
// 2: 92.3 to 293.1 mm/s
16:34:19 
// 3: 299.4 to 304.1 mm/s
16:34:19 
// 4: 310.4 to 400.0 mm/s
16:34:19 
// Lowest vibrations angles (9 ranges sorted from best to worse):
16:34:19 
// 1: 307.9° to 322.4° (mean vibrations energy: 42.67% of max)
16:34:19 
// 2: 128.2° to 142.7° (mean vibrations energy: 42.69% of max)
16:34:19 
// 3: 37.6° to 53.1° (mean vibrations energy: 46.54% of max)
16:34:19 
// 4: 217.3° to 233.3° (mean vibrations energy: 47.22% of max)
16:34:19 
// 5: 0.0° to 21.0° (mean vibrations energy: 63.22% of max)
16:34:19 
// 6: 165.7° to 201.3° (mean vibrations energy: 63.36% of max)
16:34:19 
// 7: 249.3° to 284.9° (mean vibrations energy: 63.36% of max)
16:34:19 
// 8: 69.1° to 105.1° (mean vibrations energy: 63.43% of max)
16:34:19 
// 9: 345.5° to 360.0° (mean vibrations energy: 63.57% of max)
16:34:19 
// Warning: motors have different TMC configurations!
// {'registers': {'CHOPCONF': {'intpol': '0'}}}
16:34:19 
// Motors have a main resonant frequency at 198.1Hz with an estimated damping ratio of 0.072

@slender epoch There is warning message motors have different TMC configurations!

Here is my tmc config, the x/y use autotune, z not use.

[slender epoch]

not sure what that warning message is about. @solemn ember do you know about this warning message

[quasi carbon]

How about my matrix? How to tune my printer?

[eternal lark]

can u provide us a klippy log, but rollover first please before 🙂

[eternal lark]

please keep in mind, this channel is for basic help, as the channel name states #1308016330326413393 🙂

[eternal lark]

honestly i would consider the cube quite good. PA and EM is off. must be tuned

but thats something for #slicers_and_print_help

[solemn ember]

That's a Frix question, but I can see what it is complaining about... X has interpolate: true and Y has interpolate: false. @quasi carbon you need to pick one, and I recommend interpolate: true on both.

[quasi carbon]

So on the y face’s resonance is not so bad?

[eternal lark]

Im not 100% sure its resonance

Ask in #slicers_and_print_help , see what they say

[slender epoch]

I do not believe what you are seeing is resonance.
I believe what you are seeing is PA/EM

the reason I state this is because the pattern in your lower left circle does not correspond with the shape of the letter Y

it appears to be going > > > and not < < <

[eternal lark]

its been 13 days.
any update for us?

[eternal lark]

do u need further help from us?
how do u wish to proceed with your post?

[quasi carbon]

Thank you for your help. Now my printer is running well. You can close the topic.

If I want to increase the acceleration, what should I improve?

[eternal lark]

Same as with FPV drones
Reduce weight
Stronger motors

Welcome enjoy your printer

[quasi carbon]

ldo-42sth48-2504ac and stock gantry,
what is the best result(speed and acceleration) that can be achieved?

[eternal lark]

Depends on a lot of factors

[eternal lark]

Get https://www.omc-stepperonline.com/de/nema-17-hochtemperatur-schrittmotor-55ncm-77-93oz-in-55mm-rundwelle-isolationsklasse-h-180-c-17hs19-2504s-h-v1 if u can, they are better and cheaper

U can set the website to mandarin or english

Lighter toolhead helps

A4T toolhead

[quasi carbon]

How about xol, I'm thinking about it recently.

[eternal lark]

A4T has better cooling