#Cant start klippain

191 messages · Page 1 of 1 (latest)

mystic olive
#

Hi. Just updated mmmy Klipper machine from version 11 to V12. Initially it has Klippain running. After an update I cannot start it.. Klippain was updated to, but Unknown command:"COMPARE_BELTS_RESPONSES".
As per recommendation fromm Reth, I have deleted folder of Shakentune from the Pi and have installed once again. The result did not change - it still does not know any of the commmands Im trying to use for input shaper, like COMPARE_BELTS_RESPONSES or AXES_SHAPER_CALIBRATION.
Before an update I remember having the dedicated macroses for input shaper commands, however they are gone too.
How do I fix this?

kindred star
#

hello

#

Go into mainsail and see if there are new macros for shaketune

#

also is that graph a new graph or an old graph

mystic olive
#

Update. I have managed to start belt compare. It had created the .png file for me. Yet, for some reason I cannot open it from witheen Klipper. Older files, which were created by the old version of Klippain are, however opening and displayed correctly. The only way of seeing the pictures is to download from the Pi and open on a computer. the graph linked was from the new test

#

Also, no macro (related to shekentune) appears in the list

#

do I need to update any software in the Pi itself foor a newer version to see the graphs inside Klipper? having to login through SSH to download files to a PC is an an extra step(

kindred star
#

Try changing browsers or clearing browser cache

mystic olive
#

Im inside the Orca slicer. Anyways, disregard the non working PNG files in orca - I can see them on my PC. I went through belt comparison tool and had to tighten one of the belts by a quarter of a turn and then run an input shaper tool and got a nice result. Now Im on the stage of running CREATE_VIBRATIONS_PROFILE However cannoot complete this test - giving me the error. Tried twice andd stops at the same time

kindred star
#

I believe @lavish loom has an open issue on his github about this.
Are you using a CB1 or a Pi?

kindred star
#

add cooling to it.

#

there have been reports that they overheat during this test

#

add a heatsink and cooling

mystic olive
#

Oh. I'll try this. thank you

dim acorn
mystic olive
# dim acorn Any update for us?

sorry for waiting - been away for work reasons and could not attend to my machine. I have managed to start klippain, however I cannot open pictures generated by the latest version (old version were opening within Klipper environment just fine). Also, I can do compare belts and AXES_SHAPER_CALIBRATION. However vibrations profile does not work for me - comes with error, which I mentioned prior. Probably sommething to do with the updated Klipper version, which apparently has some bugs. Once newer version will be available I'll update and try again. meanwhile this one can be closed I believe. Thank you for your timme and patience

kindred star
#

Try a different browser

mystic olive
#

this is what i have now. updated klipper

#

This happens after inputting commmmand for belt commparision

#

I did not re-flash mmy toolhead nor MCU with the updated klipper. Not sure if it is required, since machine is printing well and no issues other than klippain

kindred star
#

See if the macro named changed.
Go to moonraker and find the macro submenu

#

Then look for a second belt test

mystic olive
#

Im sorry for being slow. Do I need to go to moonraker.conf? Or did you mean to go into mmainsail settings and check the macroses there? In the mainsail i now have two sets of macroses COMPARE_BELTS_RESPONSES and compare_belts_responses and few more with and without capital letters. I've tried iinputing both ways - regardless of capital or not letters used - I am getting erroor

#

@kindred starCould you please clarify a bit? Im not sure where to look for the macro submenu in moonraker

kindred star
#

one sec

mystic olive
kindred star
#

Okay you got there

mystic olive
#

What amm I looking for here?

kindred star
#

look for two compare_belt_responses. One may be in upper and the other lower case

#

are you in group view?

mystic olive
kindred star
#

if so look down into available macros

mystic olive
#

once I input any of klippain macro - it will error out my klipper

kindred star
#

can you post your klippy.log and moonracker.log please

mystic olive
#

give me a sec

kindred star
mystic olive
#

Thanks for this clarification. I did as you said - deleted everything and set new grooup with macroses. however no change

kindred star
mystic olive
#

I have used it before the Klipper update a few days ago. Could new Klipper update brake something with shaketune? Also, I have Pi and toolhead/MCU have different versions of Klipper - would it impact this?

kindred star
#

can you set your belt shaper macro like my picture and see what happens when you hit send

kindred star
#

klipper added a new feature which will allow you to sweep the input hsaper.

#

this new feature breaks shaketune because shaketune does not have the "flag" it is looking for

mystic olive
#

Oh. so I need to wait for the Shaketune update, right?

kindred star
#

yes

#

or you can use klippers built in test

mystic olive
#

oh, how would I do that?

kindred star
#

I have the commands in this pdf

#

They are also contained in this portion of the klipper documentation

mystic olive
#

will it be comparable to shaketune or there is no comparision? I am in process of changine the motion on my printer and was hoping to see the difference betwin different parts imm uograding)

kindred star
#

shaketune is better, but that is not to say that the klipper builtin are crap

#

shaketune gives more information that is usefull in troubleshooting than klipper.

#

I talk about the differences in that PDF above

mystic olive
#

Klipper test apparently working

#

Is it how it should be? Klippain keeps toolhead was almost stationary while performing the test

kindred star
#

Yup. That is the belt test

mystic olive
#

Reth, any idea when you guys will have klippain working with the current Klipper build?

kindred star
#

Frix is on vacation. He does the shaketune

#

He is aware of the issue, and will fix it when he returns to reality.

#

But he lives in France, so vacation is more than a weekend

mystic olive
#

@kindred starone last question mate. In your PDF you wrote : "~/klipper/scripts/graph_accelerometer.py -c /tmp/raw_data_axis*.csv -o /tmp/resonances.png
(ssh)
• Files in /tmp Where do I find the tmp folder?

#

I run the script in ssh already it has finished

kindred star
#

cd /

#

It is on the root directory

mystic olive
#

doesnt look like i have it

kindred star
#

Click the ..

#

Above cache

kindred star
mystic olive
#

Oh I see. Thank you mate. Thank you very much

kindred star
#

Welcome

mystic olive
#

yeh. the internal klipper resonance tester tells me almost nothing... So far I only changed plastic TAP R8 for Stealthchanger shuttle. The rest remains stock.

#

this was the stock with R8 TAP prior to klipper update

kindred star
#

What is your absolute belt tension and how are you measuring it

mystic olive
#

With this. Im having both belts at about 2.1 on the scale. Gauge is assembled with recommended on github parts (spring size/gauge) and before changing TAP for stealthchanger I had beautifuly setup machine with the help of this exact belt gauge

mystic olive
#

@kindred starshould I try to return TAP back to see if graphs will get back to good state? - this way I will know stealthchanger shuttle is a bad idea.

#

On the other hand I saw your screenshot yesterday and you are clearly using multi toolhead mamchine - so you've made it working... What type is your shuttle? tapchanger/stealthchanger/misschanger or something else?

mystic olive
#

its a first time I see a suggestion to use 3Hump ei shaper on any of my printers)

kindred star
mystic olive
kindred star
#

That 125 hz is going to be the shuttle

#

125 is generally associated with toolhead/tap

mystic olive
#

@kindred starMate, r u using toolchanger? which shuttle did you go with and which mmmaterial is it made of? Did you get samme peak on the graph due to shyuttle too?

kindred star
#

I made it out of abs

#

I did not get the same peak. You may need to adjust the screws so they are flush/connecting with the magnets

chilly drift
#

BTW: I have the same issue, running the macros crashes klipper

#
Internal error on command:"AXES_SHAPER_CALIBRATION"
Internal Error on WebRequest: gcode/script
Traceback (most recent call last):
  File "/home/biqu/klipper/klippy/webhooks.py", line 256, in _process_request
    func(web_request)
  File "/home/biqu/klipper/klippy/webhooks.py", line 436, in _handle_script
    self.gcode.run_script(web_request.get_str('script'))
  File "/home/biqu/klipper/klippy/gcode.py", line 230, in run_script
    self._process_commands(script.split('\n'), need_ack=False)
  File "/home/biqu/klipper/klippy/gcode.py", line 212, in _process_commands
    handler(gcmd)
  File "/home/biqu/klipper/klippy/gcode.py", line 140, in <lambda>
    func = lambda params: origfunc(self._get_extended_params(params))
  File "/home/biqu/klipper/klippy/extras/gcode_macro.py", line 189, in cmd
    self.template.run_gcode_from_command(kwparams)
  File "/home/biqu/klipper/klippy/extras/gcode_macro.py", line 68, in run_gcode_from_command
    self.gcode.run_script_from_command(self.render(context))
  File "/home/biqu/klipper/klippy/gcode.py", line 227, in run_script_from_command
    self._process_commands(script.split('\n'), need_ack=False)
  File "/home/biqu/klipper/klippy/gcode.py", line 212, in _process_commands
    handler(gcmd)
  File "/home/biqu/klipper/klippy/gcode.py", line 140, in <lambda>
    func = lambda params: origfunc(self._get_extended_params(params))
  File "/home/biqu/klipper/klippy/extras/shaketune/shaketune.py", line 173, in cmd_AXES_SHAPER_CALIBRATION
    axes_shaper_calibration(gcmd, self._pconfig, st_process)
  File "/home/biqu/klipper/klippy/extras/shaketune/commands/axes_shaper_calibration.py", line 26, in axes_shaper_calibration
    min_freq = gcmd.get_float('FREQ_START', default=res_tester.test.min_freq, minval=1)
AttributeError: 'ResonanceTester' object has no attribute 'test'
mystic olive
lavish loom
#

I'm back in France and will look at it very soon. Hopefully an update will be pushed in a day or two 🙂

kindred star
lavish loom
mystic olive
#

@kindred starThank you for the council. after you suggested to adjust the lower two screws on the stealthchanger shuttle V1.1 - I massively immmproved shaper graphs.

#

@kindred starShould I be celebrating victory or is better can be achieved on 300mm machine? Could you share how your toolchanging graph looks like?

kindred star
#

I would print a voron cube using ei with 4k accel
Voron cube - 3 perimeters/bottom, 0 top/infill

mystic olive
#

@kindred starei on X or on Y axis?

kindred star
#

Both

mystic olive
#

@kindred starI've noticed one thing - the graph is 1e4 scale, so my vibration energy is like 40k on X and 35k on Y. Do I get it right - it just means my belts are undertentioned and I should pick up the slack untill graph will indicate energy as per klippain wiki in the range 500k - 2000k ?

#

BTW I've replaced stock aluminium profile on mmy 300mmmm V2 for lightweight Alu gantry and this is what I've got

#

nothing else was changed, even belts tension was preserved

#

Like no results from the lightweight gantry whatsoever. 67 AUD and some timme wasted for no reason)

kindred star
#

Can you add new pictures of the new setup please

#

These graphs do not tell you about the belts per se.

#

that woudl in klipper test custom axis. However, even this test only tells you realtive belt comparison, and not absoulte belt tension. To test absolute belt tension you woudl need the tension meter or equivalent

mystic olive
#

@kindred star thanks for coming back so quick

kindred star
#

What does the tension tool say your tension is

mystic olive
#

@kindred starits as per picture. the gauge is sort of calibrated.

kindred star
#

Increase that tension to 2.0 on both
Why 2.0 - it is a nice round number about in the middle of the tension range

mystic olive
#

This is after increasing tension up to 2 on the gauge scale.

#

So in total I decreqased X axis weight by 240 gramms, and my input shaper graphs remain largely unchanged

#

Correction. graphs actually got worse o_O

#

before changes recommmended accelerations were 6400/4700, while post changes its now - 7300/4400

kindred star
mystic olive
#

@kindred starthis one is

kindred star
#

I would use ei for both x and y.
I have forgotten where we are at? What was your question

mystic olive
# kindred star I would use ei for both x and y. I have forgotten where we are at? What was your...

I suppose I wish to know if this printer is working correctly for its size ) its V2 300mm with the input shaper graphs linked previously. Lightweight X gantry, 24V, X rail is Z1 preload, the rest are Z0, Gates belts, Stepperonline motors, HiWin rails. So if you suggesting to use Ei shaper - it means this printer is only capable of accelerations of no more than 3900 (since its the recommended Acceleration on Y). Like rly? This cost like twice the price of some others core xy printers, which come pre-assembled and have 12000 ACC on both axis. How do I improve mine, @kindred star?

mystic olive
kindred star
#

It is in the lower range.

#

you should be able to get 4k, but you have stuff going on in your y-graph which may be lower the acel

#

after about a 100 hours or so of use, I would take a second look at belt tension

mystic olive
#

This machine is 400 hrs in. Thats why seeing creality and bambu's new printers with 10k+ acc I decided to look at it again. So I've bough lightened X gantry and will transition to Monolith gantry mod. Was hoping to see somewhere close to 10k after installing lightweight X beam, but I literally see no improvement whatsoever. Questioning if I even should now proceed with monolith or it will not affect my accelerations too.

kindred star
#

Those numbers are not real for creality or Bambu. You can get the same numbers on a voron for max accel of the system.
Take a look at a Bambi profile in orca slicer.

#

Monolith may improve your accels.
It all comes down to build quality

#

Back to accels. I have 4 vorons printing 24/7. I think my max perimeter accel is like 4k. Why, because I want the prints to come out good everytime.

#

I also do toys for tots for IC3D.
Some of the people printing toys use bambu's. I will tell you that my prints exceed theirs in quality.

#

It all comes down to what print quality do you want. If you want good print quality you go quote slow (4k ish).
Generally, the faster you go the less print quality you will get.
You can do things to change this, like AWD, but standard 2WD, you need to start doing 24/7 printing tricks.

mystic olive
#

Im almost finished Monolith now. takes slow due to real life being in the way) Need to figure out what to do with the Z chain. it has to be horizontal instead of vertical (in standart setup), but myu chain bend radius is to big - does not fit betwin the gantry and the deck

mystic olive
#

@kindred starHi mate. this graphs are: left is stock v2 300 with lightwtight x-gantry and right column is same machine converted to Monolith gantry mod. belts are the same gates power grip GT, but 80 mm shorter, both motors have double shear supports, all mmotion plastic parts are ASA-CF. So after all of this my printer is barely any better than stock. Amm I doing something wrong here? With your expertise for sure you can advise me something on how to improve this. I am sure I am doing something wrong either mechanically or electrically, which does not allow this machine to reveal its potential 🙂 my old bed slinger 220 mm can do 4900 acc on x/y, so surely Voron can go faster, but I don't know what else I can do really here(

kindred star
#

You are using monolith 2wd, not awd. I woukd expect about the same performance.
@compact wedge is there much of a performance difference between standard voron and monolith 2wd? If do about how much?

#

@mystic olive What do your belts graph look like?

#

And, what is your belt tension at?

mystic olive
#

@kindred starcant show belts graph since klippain does not work on the latest klipper, from what I understand( unless you teach me how to get it through klippers inbuilt tools?

kindred star
mystic olive
kindred star
#

Make them 2.0

mystic olive
#

Ok, give me 2 min. I'll get through this and come back to you

kindred star
#

Run both test resonance test - one after the other

#

Then run this command ~/klipper/scripts/graph_accelerometer.py -c /tmp/raw_data_axis*.csv -o /tmp/resonances.png

mystic olive
#

@kindred starPerformance expectations of Monolith:
2WD: Due to the 10-15cm shorter belt path, with stiffer and lighter XY-joints the printer should perform similarly to a 50mm smaller one equipped with the same X-axis setup.

#

however I did not see 10 cm shortening - it was only 8

#

@kindred starthis is my belts graph. does 2 spikes mean beelts aren't tensioned evenly or what?

kindred star
#

generally we do not see dual peaks so close together.
This may be a 2wd monolith fingerprint.

#

let's see what @compact wedge has to say

mystic olive
#

ok, I'll wait for the guru to chime in

compact wedge
#

Or with anything

compact wedge
#

You could have 12mm AWD with a full 4040 frame and metal gantry parts, but this beam paired with SB would make the printer perform worse than the stock setup.

kindred star
compact wedge
#

There's just less you can get away the more you're asking of the same 2020 based X axis.

kindred star
compact wedge
#

but 2WD vs 2WD I think hartk's gain on his 350 was the best I've seen.

#

almost doubled on both axes

#

but for some it's about 10%

#

X is usually more limited with the stock XY joints.

dim acorn
mystic olive
compact wedge
kindred star
compact wedge
kindred star
compact wedge
#

here's the list

kindred star
#

thanks

rich canyon
#

Any idea what causes this?

Internal Error on WebRequest: gcode/script
Traceback (most recent call last):
  File "/home/pi/klipper/klippy/webhooks.py", line 268, in _process_request
    func(web_request)
  File "/home/pi/klipper/klippy/webhooks.py", line 448, in _handle_script
    self.gcode.run_script(web_request.get_str('script'))
  File "/home/pi/klipper/klippy/gcode.py", line 230, in run_script
    self._process_commands(script.split('\n'), need_ack=False)
  File "/home/pi/klipper/klippy/gcode.py", line 212, in _process_commands
    handler(gcmd)
  File "/home/pi/klipper/klippy/gcode.py", line 140, in <lambda>
    func = lambda params: origfunc(self._get_extended_params(params))
  File "/home/pi/klipper/klippy/extras/gcode_macro.py", line 195, in cmd
    self.template.run_gcode_from_command(kwparams)
  File "/home/pi/klipper/klippy/extras/gcode_macro.py", line 74, in run_gcode_from_command
    self.gcode.run_script_from_command(self.render(context))
  File "/home/pi/klipper/klippy/gcode.py", line 227, in run_script_from_command
    self._process_commands(script.split('\n'), need_ack=False)
  File "/home/pi/klipper/klippy/gcode.py", line 212, in _process_commands
    handler(gcmd)
  File "/home/pi/klipper/klippy/gcode.py", line 140, in <lambda>
    func = lambda params: origfunc(self._get_extended_params(params))
  File "/home/pi/klipper/klippy/extras/shaketune/shaketune.py", line 173, in cmd_AXES_SHAPER_CALIBRATION
    axes_shaper_calibration(gcmd, self._pconfig, st_process)
  File "/home/pi/klipper/klippy/extras/shaketune/commands/axes_shaper_calibration.py", line 26, in axes_shaper_calibration
    min_freq = gcmd.get_float('FREQ_START', default=res_tester.test.min_freq, minval=1)
AttributeError: 'ResonanceTester' object has no attribute 'test'
kindred star
#

Try restarting printer

rich canyon
#

Still a problem, but I just realized this thread is for another issue. Going to start my own post