Provok 3D Nozzle ADXL
I was trying to set up my nozzle ADXL that I had just got, and I put it on my printer and tried to run the built in input shaper macro in RATOS 2.1.0 but i continued to get an error about an unknown probe. Does anyone know how to modify the macro / fix this error? The error was "Invalid adxl345 id (got ff vs e5)."
Solution:Jump to solution
You need to name the adxl345, and you need to reference that name in your resonance_tester configuration. You also need to specify accel_chip_x and accel_chip_y and not
accel_chip200 Replies
That error almost always means it is miswired. Triple check the order of the wires and all the connections
I am using a USB adxl
Then the configuration is likely wrong. Share your ratos.cfg and printer.cfg 🙂
I'm still inexperienced with the 2.1 ratos.cfg. Just to confirm you haven't edited the ratos.cfg right?
i have not edited it
And youy re-ran the configurator to add the adxl?
uhhhhhh
no
software-wise, what did you do to add it then?
nothing
i would re-run the configurator
doing it now
would I click wired to host computer?
if I have it pluged into the pi?
yeah. Host computer is the raspi, mcu/mainboard is the octopus
ok im going to go test it now
I got the same error
Interesting. I hope it doesn't indicate the adxl itself is bad :/
When I plug it in the light turns on so
I think it might be something with the config not being meant for USB adxls
not sure how to fix it though
the led could just be indicating power is to the board
Oh, does the configurator not distinguish between usb and not usb?
it does not
That strikes me as odd. Tagging in @blacksmithforlife
This isn't a common thing to do, I think most are either using the accelerometers on a controlboard like the EBB42, or the one in the beacon. Not a completely separate usb adxl so it might not be configurator-able. In which case you'll need to set it up manually. Did your device come with any instructions?
It did not come with any instructions
I can send the link do the ADXL if you would like
nah, I'd just start googling how to set up a USB adxl in klipper. Ratos is just klipper with some extras on top. So any setup instructions for klipper should work fine.
ok
from my understanding you need to first flash the board that has the ADXL on it. If it isn't supported by klipper you are out of luck. Second is you have to specify the ADXL you want to use otherwise it will default to the main MCU one (which isn't connected in this case and is why you are getting a bad reading)
I was just following a tutorial, and I added the config file for the ADXL and I got this error.
The board supposedly comes preflashed
I can try flashing it again though
Do I flash it while it is plugged into the printer?
because the documentation doesnt specifiy.
https://docs.provok3d.com/nozzle-adxl-v13/#2-klipper-config
Also it says plug into SBC but I dont know what that is
Single Board Computer, the raspi
Ok
so do I just run those two commands once I ssh into the rasbery pi?
yeah
Ok
You'll still need to add the lines spelled out in the cfg on that page
ok
ok
and I made a accelerometer.cfg script
with this in it
and then I do this right?
Because that is what it said in the video that I was watching on how to setup a USB adxl.
Seems right, id give it a shot
Now I got this error
At least we are making progress though
its a different error from before
I am also going to assume that this will also happen for "shaper_type_y"
I'd assume so yeah
Any suggestions on how to fix?
share your updated printer.cfg now?
This is the only thing I changed.
your
[input_shaper] is commented out, but the lines below it are uncommented. Put a # back in from of each of those shaper bits and try againok
that error message is saying the section "gcode_macro RatOS" shouldn't have stuff about shaper_type_x. Notice that thats the last valid section header
Im back to this now
If I comment out all of those lines
than I dont get an error
Perfect, can you run the test now?
it doesnt work
I think I may have to flash the firmware again
i think i need to put the adxl into boot mode
You've got some things to try and levers to pull, i'm sure you'll get it soon. I gotta run for now but good luck 🙂
I tried re-flashing the firmware and messing around with the config, and it still doesn't work.
Do you think it is worth it to just get a non usb adxl?
and just mount it to the toolhead?
What's the name of the board you bought? Maybe it is already supported by Ratos
Provok3D Nozzle ADXL
Provok3d
Provok3d Nozzle Adxl
Hard mounting a adxl directly onto the surface of the test object provides a more accurate, sensitive, and stable measurement compared to plastic mounting.
ok, and does it show up in the configurator for flashing?
in the RATOS configurator?
yes
wdym does it show up
I am not convinced the board is flashed with klipper firmware. We need to make sure that is correct before moving on
The configurator makes it easy to flash boards if it is supported
this is what shows up
that is for after you have flashed the board
I flashed the board with the instructions from provok3d
it says it uses Ampon
and does klipper understand that?
I would assume so
why would you assume so?
Because in the pictures on the product page is shows it on a voron
It also says it on the github
Solution
You need to name the adxl345, and you need to reference that name in your resonance_tester configuration. You also need to specify accel_chip_x and accel_chip_y and not
accel_chipie
If you search for my name and usbadxl you'll find me repeating this quite a bit 😂
Can you explain what config this would go into?
always into the printer.cfg, don't edit other files.
As @TheTik said, you never ever modify anything other than printer.cfg or any files you may have added yourself.
Well moonraker.conf if you wanna do stuff to moonraker, but that's it 😄
Do I still add [include nozzle-adxl-v13.cfg] to printer.cfg?
Depends if the config is all borked
yeah this needs the changes i listed
You can still keep it in that file and include it
instead of putting it directly in printer.cfg (that's effectively what the [include ...] section is doing)
so would this be right?
remove
probe_pointsOk
And then will the built in macro in Ratos work?
Yep 🙂
Thanks, I'm going to try it now!
One last question, once it gives me the recommended input shaper, how do put that into the printer?
like where do i add it in the config.
I thought it puts them in the Save_config section automatically. Once it is done, see if it shows up there
where is the save_config section?
Its the last part of my printer.cfg but I'm stil on 2.0. That might have been changed.
Your printer.cfg has a line about this though:
You can run SHAPER_CALIBRATE to automatically calibrate your input shaper configuration, if you just want to get started.so run that command, let it finish, do whatever it says (might be something like SAVE_CONFIG and restart)
Ok, Thanks!
Just realized I only ran the macro the generate graphs.
that's fine, that's usually prefered so you can put in MZV instead of ZV (which is shit).
Is this fine though?
You'd do that like so:
[input_shaper]
No you should definitely run the belt tension macro and tune your belts, those are all out of whack
Should I delete that?
Or should I just run the belt tension macro now
I would, it will override anything you put in manually
Ok
it won't impact the test, but your manual settings will be ignored as long as it's there.
Ok I have the graph
what do I do with it?
Is this a v-core 4?
No
What is it?
V-core 3.1
Ok, same principle, go read the V-core 4 commisioning docs, they'll explain the process 🙂
Specifically this part: https://docs.ratrig.com/commissioning-guides/v-core-4#belt-tension
Does this graph mean that Both my belts are rubbing up against something?
Also how do I know which belt is which, and to loosen or tighten either?
It means you have something oscillating at roughly 125hz that shouldn't be oscillating.
could It just be that my ADXL is loose?
Those are your main because
They're unevenly tensioned, and they don't have the same length from attachment point to attachment point.
So should I tension the upper belt and run it again?
Yeah it could easily be a poorly mounted ADXL, it could also be your hotend fan spinning if it's badly balanced.
you should locate the source of the 125hz vibration first
And how exactly would I do that?
check for loose parts
proper wiring management etc is important
everything should be tight and secure
If you're already at that point, you can use the realtime analysis tool to pin point the source. Start the oscillator at any frequency that causes a peak at that 125hz frequency, then touch suspected parts on your printer until it goes away, then you've found your source.
Whenever I try and start streaming the data to the realtime analysis tool, I get this error.
And when I switch it to host I get this error.
gah shit i forget you're using the usb adxl
lol
We don't support those
Sorry!
Ahh shucks
Wait.. I guess you could trick it..
How?
If you name your adxl
rpi instead of usbaccel it might worklemme try
Like this?
or change all of it
Yes except you need to use
rpi instead of usbaccel in the resonance tester section as wellI think it worked!!!
Should I oscillate a or b belt though?
seems like it's most obvious on A
You can switch between them to check
Whenever I hold the toolhead, the vibration goes down.
Yeah that's kind of cheating 😄
It's supposed to measure the resonance of the toolhead. The graph represents all resonance recorded at the toolhead, if you hold the toolhead all the resonance goes away 😄
You just want to kill the 125Hz peak
Also keep in mind that you most likely need to oscillate at a very different frequency than the resonance frequency.
I was at 44Hz
What does the graph look like?
you only need to lightly touch parts of the machine with a finger to impact the resonance
Yeah this should be good enough (44hz)
to track it down
although it looks like it's quite a bit higher in frequency here
Should I be touching things on the toolhead? Or all around the printer
Just went back to check, it's 130hz so not too far off.
Everywhere
Try the idler stacks for example
It can even be a fan attached to your electronics on the back
It'll pick up a clap in the next room, it's quite sensitive.
if you have it standing next to anything that might vibrate that will do it too
And since it's such a high amplitude i suspect it might be something external / always running.
uhhhh
Can you see the 130hz peak without starting the oscilator?
I may have it next to another printer that is currently printing
Could be that.
I would expect much lower frequency resonance from that though
But it depends on a lot of things
Try setting the oscillator to 0hz (that disables it)
If you still see the 130hz peak, it's not from your motion system 😄
This is without the oscillator
Well, it's deifnitely generated by your v-core
Is there a way I could set my hotend fan to 0%
It should not be running if your heater isn't on
oh your right
im just blind
If I hold the umbilical cord than the vibration pretty much stops
Found your source! 🙂
So how would I go about fixing that
or should I just not worry about it
And going back to before, should I just tension the top belt?
Depends on your umbilical, can you show me a picture of it and the attachments at both ends?
Once you've fixed the 130hz issue, you should tension the top belt, yes. However, be careful you're not racking your gantry. If you followed best practices you would have tensioned the belts while checking the gantry squareness against the front of the printer. When it's square, the tension is equal, if your peaks don't line up at that point that means your belt lengths aren't equal.
How would I fix the 130hz issue though?
Like what would I do to the cables?
Right there seems to be a bunch of slack in the middle that's drooping over, if it's already as short as it can be, there are different ways to deal with this, piano wire (spring steel round wire) is one, but it can also cause similar issues if it's too thick.
I was thinking of getting the RR toolhead
should I just wait to get that
Because that supports a toolboard
and then I wont have to worry about it
Won't change your umbilical, but it does improve your recommended shaper acceleration
It's a good deal stiffer than EVA
And it's better balanced
Why wont it change the umbilical?
same wires.
It would make it only 1 cable and the ptfe tube
Are you talking about a toolboard? you don't need the rat rig toolhead for that.
But yes, that helps
there is no space to mount it on the EVA one
There is plenty of space 😄
It replaces the grabber between the LGX motor and the 4028
There are multiple mods for that
Ok, I am still going to upgrade to the RR toolhead
and get a Orbiter 2.5
so I will finish this once I get that
Fair enough! 🙂
anyways, when I take the accelerometer off, do I need to change the config at all?
It is better mechanically, but the ducts aren't as good as EVA
Its fine, the fan I have is so overpowered it can push enough air through
(worse cooling performance).. With a 4028 that mostly just means louder fans though as you need to run them at higher PWM.
yeah exactly 🙂
I am debating weather or not to replace it with a 5000rpm noctua
that will do absolute nothing
will it make it quieter?
do not ever consider noctua for any situation that requires pressure
they suck ass
ok
yes and move no air at all
lol
because the one I have now is 23000rpm
and its REALLY loud
You don't need to run them at speeds where the noise is a problem on EVA3 though
i rarely ran them faster than 45%
lemme check what mine is limited at
on the RR toolhead it needs quite a bit more than that
It depends a lot on the fan
my fan is limited at 60%
I just relized my my min layer time is 100s
jesus christ 😂
I just made it 25
because my hotend is so slow
especially with an lgx lite
it still prints really slow though
40min benchy ðŸ˜
Also do I have to change the config
when I unplug the adxl?
then limit your flow rate in the slicer and not your layer time.
yes, part of why i hate them. DangerKlipper has non-critical mcu's for this, which is going to make them a whole lot more useful.
What do I have to change in the config? Just comment out the line in printer.cfg?
indeed 🙂
Do you just work on RATOS or the entire printer?
Both, but RatOS is my baby
Ok, I was wondering if you would be able to tell me if the RR V-core 3.1 toolhead supports the toolboard that mounts to the back of the orbiter 2.0
Like if it would fit
It does, but i wouldn't mount it like that, i would use a proper holder and i'd get the 42 version, it's much better organized and provides much better mounting.
what is the 42 version
so usually those toolboards come in "36" and "42" models, because they're sized for nema14's (36mm) and nema17's (42mm)
ok
cheapest option currently is the EBB 42 1.2
I cant find anywhere to buy the 42mm one though
Because I want a usb one
not can
oh nvm
im stupid
i can just guy it from ratrig
because it says usb/can
Do you know if the hotend mount is the same as the beta?
because I found an adapter for revo
but it is for the BETA 2
version
they're all usb
You can't get CAN firmware on there without a usb connection anyway. Unless they shipped with Katapult, and there are not many that do.
CAN is the "extra", not USB 🙂
Do you know the answer to this?