First Input shaper test on V-Core 3.1 500, looks horrible
Can anybody help me to decipher my resonance test?
I know that it looks horrible, but from the graphs shown, what am I supposed to do to make it better?
Which issues are present?
290 Replies
other-emeraldโข2y ago
Start with a belt graph so we know whats going on there
How do I make that?
1e3 range means you aren't picking up anything from the adxl
What does that mean?
The range on those plots should be up to 1e4, not 1e3... 1e3 means the adxl is picking up barely any acceleration... Look at some good plots, you'll see what I mean...
So how do I get it to be 1e4?
Could that have something to do with belt tensioning? Is it the ADXL simply not work?
I'd look first at your adxl connections and how it's mounted... Something isn't right with it
Seen this video a couple of times
3D Printers & a Whiteboard
YouTube
How to Read and Analyze Input Shaper Graphs Generated by Klipper
This video is supposed to be a brief overview on what the input shaper graphs can tell us. It is the first video of a series of videos which will give a more in depth view on the kinematic system and input shaping in Klipper.
As always: feel free to jump to those parts, that help you with your problems:
00:00 Intro
00:57 What we can learn from...
And wired exactly as shown in the guide.
Looks OK physically
Not at home or I'd send a typical graph
Are your belts tight enough? Grub screws tight? Wondering if there is some backlash that results in the print head not moving as much as it should
Donโt think the belts are tight enough no, they are quite slack after starting up the machine for the first time and moving the head around. Also the printer is ontop a rolling table, which I found out was a big nono. So I think that I tomorrow will move it down to the floor and do some belt tensioning. If I still have such bad results, or even still bad results, Iโll write back.
Which grub screws should I check?
And where are they?
The grub screws on toothed gear for the x/y motors
Here is the belt tensioning graph:
other-emeraldโข2y ago
Belts look very loose,try tightening and get the graphs parallel
other-emeraldโข2y ago
Shuld look something like this
Jesus from facebook told me that the belts should be around 53-62hz on both, that what I did here.
In relation to the phone, it say both are at 56hz.. but it is still way off..
Or is that way too little?
56hz that is
other-emeraldโข2y ago
๐คทโโ๏ธ I dont tune by phone apps ๐
How then? XD
I'm very lost atm.
other-emeraldโข2y ago
ADXL tension and rerun little by little
Try tightening upper belt only and see where the graph is going
My y-axis on the graph also only goes to 175k, but yours goes to 2mil.
I saw that when tightening the belts, it goes up. Is it supposed to be much tight?
other-emeraldโข2y ago
Try get them equal first then you can play with how tight ๐
If you run them too tight you will get artifacts
Also this from facebook is at 2mil and is acceptable.
But like mine is at 175k, while so many others are up in the millions
other-emeraldโข2y ago
You will get there when its close to perfect
Donโt know if you can see anything from this, but is this way too loose?
other-emeraldโข2y ago
Not way but a little
Does the belt slap agains the frame when you run Belt_Tension?
I can see if it does.
Doing the test, the belts vibrate a lot though. Don't know if that is how it should be.
I have been going around all around the printer with a torque wrench and I found a surprising amount of screws that was quite loose. So I have tightened them as well.
other-emeraldโข2y ago
They will vibrate kinda the point ๐
The grub screws on the motors are totally loose.
It's almost running in free gear XD
So, I will be tightening these first and foremost and then trying again.
How it has even been able to move the head, I donโt understand. Will do lots of testing a tuning tomorrow. Havenโt done a test yet after fixing that problem.
other-emeraldโข2y ago
Gogo then buzz it
Wauw this looks much better after setting the pulleys correctly.. and tightening the grub screws ๐ฎ
How do you think it looks now?
Including the belt tension (which I can see can be a little better)
I have tensioned the belts to 59hz. Now the the graphs looks like this:
Is that better now or too tight?
see ^
other-emeraldโข2y ago
Try loosen button and tighten top same amount and see whats happen
Oki doke will do
Can't get closer than this:
Input shaper x and y changed a little as well. A little worse.
other-emeraldโข2y ago
Belts got closer to even tension,Id keep trying to get then alike before doing X/Y runs.
Ok, have been trying all day though...
It is very sensitive, I must say
other-emeraldโข2y ago
Yeah ๐
If you get boored do a Shaper_Calibrate and fiddle around with something else for a while and come back later if you feel the need ๐คทโโ๏ธ
Lol XD
What does that do?
other-emeraldโข2y ago
Auto input shaper
Puts all the data in your Printer.cfg
So it's not the same as this:
other-emeraldโข2y ago
Does it generate the data for printer.cfg tho? Tough it was just for graphs
Maybe, don't actually.
Don't even know how anything of it works XD
So when running that command, I will not need to do anything else?
Other than ofc. testing at what acceleration it runs the best, by printing something.
other-emeraldโข2y ago
Just save config after,it even asks you to so ao
It saved this:
Do I need to uncomment this?
other-emeraldโข2y ago
No
What does it say on top? ๐
But like, so #*# means it is not commented out or? XD
other-emeraldโข2y ago
Nooo ๐
Ok lol
So I don't need to do anything after saving the config in the command line ?
other-emeraldโข2y ago
Nope,its done deal
Ok cool
Do you know the LGX lite and how it works btw?
other-emeraldโข2y ago
Nope,using orbiter 2.0
Oki doke fair enough ๐
If work on other stuff like pressure advance for example, if continue to work with the belts afterwards, will I have to do that step over?
other-emeraldโข2y ago
If you change speed/acc yes
PA is 3min print tho,easy
True, but only used as an example here ๐
other-emeraldโข2y ago
Ill leave this here and say good luck ๐
Lol, is that yours? XD Looks pretty amazing
Daum, really want mine too look like that..
other-emeraldโข2y ago
All stock 3.1 500,then i melted all petg parts ๐
Daum oh no. Hope you had another printer that could print the ABS parts?
other-emeraldโข2y ago
My petg melted during abs printing ๐
oh XD
That's so unlucky XD
Is e-step needed to be calibrated? Can't see it anywhere in the tuning guide.
other-emeraldโข2y ago
Dubble check it,RR std config is prob spot on
Cool
Where should the be placed when tightening the belt?
I canโt in no way and for some reason get nowhere near them touching and be alike.
There is absolutely no logic to it. When tightening when, they will either go up or down and the same with loosing them..
It is totally random and makes no sense to me.
other-emeraldโข2y ago
What?
I canโt get the yellow one to go down and blue one to go up..
If I for example tighten the blue one, it goes down and the yellow one goes up. If I loosen it it will br other way around. The same if do the same with the yellow one.
other-emeraldโข2y ago
Have you checked the height of motor pulleys so they are aligned with the belt?
Does your graph still have 2 nice peaks?
If you tighten lower = blue it shuld rise,if not it can be too tight already.
Are you tightening 1 and loosening the other every time?
Some times, Iโm only loosening or tightening one of them.
Is this way too tight?
other-emeraldโข2y ago
it's on the tighter side but shuld not be a problem
How is the belt graph looking atm?
Not terrible, but not great either..
other-emeraldโข2y ago
blue is yikes ๐ค
Yep
Spent all yesterday as well. No luck either
This motor 1:
This is motor 2:
other-emeraldโข2y ago
If you bring X all they way to the front does it hit the stops at the same time both sides?
It looks like that to me, that it does.
other-emeraldโข2y ago
you need to turn motors off and push it by hand
Ah
The motors off macro does not work..
Itโs still very hard to push and it is not smooth by any means.. does come to the end at the same time though. As far as I can see.
other-emeraldโข2y ago
Give it a good push! Also sounds like you are out of square on the rails making the ratteling noise
other-emeraldโข2y ago
other-emeraldโข2y ago
Like the belts are forcing the rails to push on the railblocks
You can try loosen the 4 m3 bolts that holds XY joiners to the rails a little push it around by hand and see if its quieter
How out of square?
These on the bottom?
other-emeraldโข2y ago
Yes,bring it all the the front,printhead in the middle and loosen them a lille both sides
It can also be the middle (on the print head) one that is too tight, right? Or is that some other issue?
other-emeraldโข2y ago
Would not make noise when you only moved X then
When moving it slowly, it sounds like it comes from the middle tho
Tried to unscrew them a little, but did nothing...
other-emeraldโข2y ago
Nozzle dragging? Idk what noise that would be
Not the nozzle dragging
Moved the bed down
other-emeraldโข2y ago
Try loosen both belts 1/2and see if it get better.
Most of it is the extruder gears I think...
Pressing on them alsmost removes the sound...
other-emeraldโข2y ago
Extruder gears?
The ones sticking out
Could you show me via video how tight your belts are?
other-emeraldโข2y ago
Soon printing atm
Awww XD
Also on the X gantry.
I noticed that the upper belt is beginning to feather at the top of the belt also.
other-emeraldโข2y ago
Pulley too low?
In relation to this picture, the pulley is all the way up here:
other-emeraldโข2y ago
other-emeraldโข2y ago
@chrizzar
And how tight is it on the x-axis gantry?
Yea, I do think mine is waaaay too tight..
Mine is like a guitar string XD
other-emeraldโข2y ago
Im printing again.. ๐
Haha It's fair enough. I have all day XD
other-emeraldโข2y ago
Loosen both belts 3turns and start over then then go 1/4 at a time until ots getting worse then you loosen 1/8 at a time
Oki doke, I'll try that. I'm starting over completely XD
Should I not be able to move the print head when clicking this button?
other-emeraldโข2y ago
Yea shuld disable all motor power
It does not however.. I can not move it around when they are off.. I have to turn off the machine to do it.
Now it looks like this:
Much more loose than before.
Is the opposite belt supposed to resonate when plucked?
It does not rattle so much anymore either, after loosening them.
What does the number to the left mean?
Before it was as 3mil, now it is only 1.2mil
other-emeraldโข2y ago
How big/powerful the resonance is i think ๐คทโโ๏ธ
Start tightening them lill by lill but try keep them close
ok
other-emeraldโข2y ago
And? ๐
Have been positioning the pulleys according to the manual all this time XD
First begun adjusting now
Now the x gantry is skew
They donโt hit the front at the same time
Right have to be pressed up against the stop
other-emeraldโข2y ago
Fix that with the belts then ๐
It moves like this now..
What to do about that?
This ^
other-emeraldโข2y ago
Looks like binding,did you print all the parts yourself or are they RR?
They are printed by ffb.
Canโt print ABS myself.
other-emeraldโข2y ago
And you are sure pulleys are tight and right height?
I have tried everything to make them look like this:
But no matter what I do, they will always have either teeth at the top or the bottom.
other-emeraldโข2y ago
Well you started with this and now it looks worse and is binding so..
So this is right:
And this is left:
other-emeraldโข2y ago
That one looks like No ๐
That looks fucked yea
Can only be seen by picture apparentlyโฆ by eyesight and measurements, it should be correct..
other-emeraldโข2y ago
You sure?
Yea it is not good right?
other-emeraldโข2y ago
By the picture no
Much better
Doing it using my phone XD
Also much smoother
other-emeraldโข2y ago
to tension graphs then
Yup and I want to check the left out as well first
Might as well do it for both of them, if the left one is skew as well
Or not aligned
The left one is fine ๐
other-emeraldโข2y ago
You said that about the right one aswell
Yup, but checked doing the camera thing
It does not rattle so much either anymore
other-emeraldโข2y ago
No binding = no side loading on rails
Ah okay, I understand ๐
Thatโs the left one
other-emeraldโข2y ago
Looks ok,go graph yalla yalla
Lol looks horrible now when it is aligned XD
other-emeraldโข2y ago
Its not align but pulleys are right height ๐ซฃ tighten blue
Yup.
Yea pulleys are at the right height yes. Nothing had been done to the belts yet after this change.
Canโt believe that tightening the belts is the hardest part of the build XD
Super annoying.
I appreciate all the help from you though ๐
Getting better ๐ฎ
Do not screw this up now.. do NOT screw this up now ๐
What do you recommend me to do here?
other-emeraldโข2y ago
Run shaper graph and have a look
How tights are belts?
This tight
other-emeraldโข2y ago
Run a shaper graph
Oki doke, doing it now
Looks pretty good I think, what about you? ๐
other-emeraldโข2y ago
I would leave it as is
Its looking very nice โค๏ธ
Cool cool
OMG Thank you sooo much for all your time used and most of all, all your help and guidance! ๐๐โค๏ธ
I really appreciate it ๐ ๐
Is this the only filament and print settings the superslicer profiles comes with?
other-emeraldโข2y ago
no shuld be more
Is this the ones?
I know my presets are fucked at the moment XD
^
other-emeraldโข2y ago
yea
Ahh okay cool ๐
Guess I should start with 0.20mm NORMAL?
About to do pressure advance and such
Or does it not matter
other-emeraldโข2y ago
everything matters when doing PA
guess HF means High Flow
Which one should I go for then? XD
other-emeraldโข2y ago
Take what ever and get a feel for it.
Oki doke
One last thing for today (I think XD)
How do I enable bed leveling and bed mesh in the slicer? Or is this done automatic before every print?
other-emeraldโข2y ago
Have you installed PAM?
Nope, lol
Have to figure that one out first then
other-emeraldโข2y ago
Yea you rly want PAM on a 500 ๐
Oki doke, else it takes forever
other-emeraldโข2y ago
mhm
Do I have to enable SSH firstly on raspberry pi to install the package, or is already enabled by default by the RatOS image?
Or is entirely unnecessary to use SSH and I can simply use the command line within RatOS?
other-emeraldโข2y ago
SSH i enabled,and is the way to go
Ok cool cool
Also just to be sure, is it this that should be enabled, that is explained in this step?
What do I do about this though XD?
other-emeraldโข2y ago
IDK xD
Just had to say yes (because it's the first time) XD
@squeegiee Ok so now I have it installed, updated my printer.cfg and updated my moonraker.conf to get updates, as well as added the new start code in superslicer:
Now what? ๐
other-emeraldโข2y ago
Enable variable bed meah
other-emeraldโข2y ago
@chrizzar ๐
Noice, I will take a look ๐
Also to make it actually touch the buildplate, I should use babystep, right?
Or am I supposed to align it on the plate perfectly and set the distance to the probe for 1mm?
Also should I set the the upper one to True as well and the lower one to 400?
(is that better)
other-emeraldโข2y ago
As long as the Variable relative extrusion is On or Off in both printer cfg and Slicer its ok. Speed is just cuz why not ๐
What do you want to touch?
Make the printhead touch the buildplate
After what it has to do, it just tries to lay filament down in the air
other-emeraldโข2y ago
Adjust Z with paper first and then tune with print
"Adjust Z": With baby step or?
other-emeraldโข2y ago
Ofc,you dont want to ruin bed or nozzle
Oki doke
other-emeraldโข2y ago
Wish you luck,bedtime now but baby will be awake in 1-2h so ๐
๐ค
Okay, thanks a lot for your help ๐
๐
@squeegiee After running "SHAPER_CALIBRATE", do I need to update my printer.cfg with my results from the test?
[printer]
kinematics: corexy
max_velocity: whatever
max_accel: whatever
max_accel_to_decel: whatever
max_z_velocity: whatever
max_z_accel: whatever
Also how do I figure out what the velocities should be as well as max_z_accel?
other-emeraldโข2y ago
Just save config after,the max_ is determined by your stepper config,do you have cooling on the drivers?
Yup, I have.
One 120mm fan
other-emeraldโข2y ago
And selected the right config for that?
So I should uncomment this and do nothing else?
And outcomment the top 3 lines ofc.
other-emeraldโข2y ago
Yes please,its not and ender3 we are playing with
lol, nope
Just haven't played with klipper or mainsail before, so I know nothing about it XD
Only Marlin and Octoprint, but that is simple setup (because there is so many tutorials out there).
So basically after that is done, everything should be set up for printing at "high" speed on the server side (not in the slicer yet ofc.).
And I don't need to change anything else?
other-emeraldโข2y ago
Yea giving steppers some extra juice โค๏ธ
Cool cool, only need to do skew correction then.
Or do I need to do that first?
other-emeraldโข2y ago
No
Okai, is it something that is worth doing? (never heard of it before)
other-emeraldโข2y ago
Suppose it depends on what you are printing i have not done it myself
Oki doke. Fair.
I see, the machine panel has changed
Noice
other-emeraldโข2y ago
Have you melted any plastic yet?
haha nope, I'm running with ABS plastic parts ๐
Though I did break both of these in the first run of the z-tilt XD
So had to reprint them in PLA. So at some point I have to re-print them in ABS again, before adding the panels XD
other-emeraldโข2y ago
Mounted them too low?
Mounted them the same place as is now.
other-emeraldโข2y ago
How did you manage to ruin them then?
It drove up into them...
On the z part of the RatOS, it did also say -5 for some weird reason.
Broke them into 2 parts XD
So it can apparently not be that skew, when running z-tilt
The stepper motors are howling now, can hear they have gotten more POWER!
Much faster z-tilt test ๐ฎ
squeeeak
@squeegiee What do I do in SuperSlicer in relation to speed settings?
So the newly added power and input shaper actually can be used?.
other-emeraldโข2y ago
Idk use HF as a start and try something
It's more like, should this be disabled?
Oki doke ๐
other-emeraldโข2y ago
Shuld ๐ถโ๐ซ๏ธ๐ idk prob not as a first go
But is it not handled by RatOS?
Or is it only used for estimating the print time?
in Super Slicer
other-emeraldโข2y ago
It overrides ratsOS i think,im@not very good with software side of all of this
Oki doke, I'll try out something ๐
@squeegiee Woaw it is a very capable printer ๐ฎ
Printing at these speeds it prints very well (the quality)!
Not sure if it even reach these speeds with 5000 accel though.
I can be begin to see some round corner though.
Could this be because of pressure advance is not setup for this speed or?
other-emeraldโข2y ago
Im a slicer newb,but id look and see why its not on all corners,different speed? Cooling? Middle is much cleaner then top/bot
Oki doke, top and bottom of the calicube itself is very smooth though ๐
Also the dimensions are spot on (in the middle).
@squeegiee Do you know what le5 compared to le4 mean?
Added the panels to the printer, so I thought I would do the test again.
It changed to le5 however, and I don't know how to interpret that..
Also the belt tension looks very good now. Only need to tighten the yellow one a little bit.
other-emeraldโข2y ago
I5 better then I4 ๐
Uhh noice
Also is the bed leveling system and PAM not supposed to conquer problems like this? :
But if I go any higher it looks like this:
Also looks fine in the other side..
other-emeraldโข2y ago
It is,try make a post in fix my print. As said before Slicer aint my strong side ๐๐
But wider and higher first layer to get more squishy in the groves would be my first test. That way you can have a lill higher Z and maybe get by.
How is your mesh looking with pam @ that print?
How do I see that?
other-emeraldโข2y ago
Height map in Ratos.local
@chrizzar
Ah okay, will try that when the print is done (in about 30 mins +-)
other-emeraldโข2y ago
You can see it while printing ๐คทโโ๏ธ
oh I can lol
other-emeraldโข2y ago
How much do you have overall on the bed? That looks like alot
That I can first see when the print is done though, right?
Yesterday, it was not much.. because of the corner it was like 0.4, but everything else was nice.
As said, because of that one corner, it's fcked XD
other-emeraldโข2y ago
Fix that ๐
How though
From the making measurement with a laser, it is square all around XD
other-emeraldโข2y ago
Say that to the mesh ๐
lol XD
How do I make it better though?
other-emeraldโข2y ago
Awww that was the one I used when assembling the printer... Did not do much then XD
@squeegiee In relation to the video, do I then need to push the gantry up here?
At A.
Since, we don't want to (or is hard to) move D, where the motor is
other-emeraldโข2y ago
A up or B down @chrizzar
I literally can't get it better than this:
other-emeraldโข2y ago
Its heck a lot better then before ๐ฆ
True true
other-emeraldโข2y ago
Melted any plastic yet? ๐
Yup, I have ๐
But only the stuff that needs to be melted ๐
This set was your best graph, I understand that the tension wasnt as high as the other person that they posted, but I have yet to get this high. Mine has been hitting 5k/11k, and your 6k/20k is very good for a 500. If you can get back here I think youll be set. If you look at the indiviual x/y the peaks are very very smooth. Seems there was a bit on the y at the end, but ive seen a lot of graphs with bumps in the 100's. I just waited all day trying to get it right and going to keep going tomorrow
Also ona Vcore 500, first printer, the quality build also affects the graphs, even a loost nut somewhere. Additionally the smaller the vcore the higher then numbers ive seen. So for a 500 thats excellent