#[SOLVED] no UUIDs Found

1 messages · Page 1 of 1 (latest)

fierce sage
#

essentially i wanted to move over from ERCFv2 with BTT_MMB to BoxTurtle with AFC lite.
at first i used a raspberry pi 5 with a dedicated U2C board with the following setup on the can network: ERCF(terminated)<-> U2C<->SB2209<->Carthographer(terminated by default) and it worked fine.
Since i wanted to change to BoxTrutle i wanted to simply swap those two. So i disconnected to ERCF and hooked up the Boxturtles AFC Board and flashed katapult via USB. No issues until i tried to check die unused UUID and i got nothing in response.

I checkt all the connections but they seem to be fine. 120Ohm terminating both ends and i messured 0 ohms across CAN_H and 0 ohms across CAN_L. 60 ohms across CAN_L and CAN_H. Afterwards i tried to flash katapult to my SB2209 to see if the issue was produced on the other and of the bus. now i can't get a uuid from my eb2209 aswell.
Furthermore i've red an announcement from the carthographer team noting that AFC produces interferences on the line killing the carthographer. Indeed my carthographer gets very warm and messures ~93 ohms accross CANL/H. i dont mind for now and i disconnected the probe and terminated ebb2209 so the setup is like this: AFC(terminated)<-> U2C <-> EBB2209(terminated)

i went as far and threw in another U2C module to seperate those two so i got AFC<->U2C_1(can0) and U2C_2(can1)<-> SB2209
this did not work either so i went back to single U2C. Also i'm using a RPI3 for now since i tried different distributions but there is no official release of MainsailOS for Pi5. Now i'm using bookworm lite and installed everything via kiauh
I described my issue on the voron discord beforehand.
#1076243803947667516 message
I'm out of ideas but i'm kind of sure its not the wiring...

uneven minnow
#

Can you post your katapult menuconfigs?

#

Also <@&1304550334839918672> assemble

coarse sun
#

Hold on, before we go further, you have seen the warning of running Cartographer and Turtle in same CAN network, correct?

uneven minnow
#

Yes he mentions it

coarse sun
#

Okay, I kinda skimmed and didn’t see it, let me go read more thoroughly

jovial rapids
#

I would have to guess we need to see what settings were used during the make menuconfig if you're confident your wiring is good. If you had a successful flash and have no UUID its typically bad CAN H/L connection or bad settings.

fierce sage
#

this is my current katapult config

image.png
coarse sun
#

That’s incorrect for AFC iirc

fierce sage
coarse sun
#

It does

#

another thing, is your CAN network speed actually 1000000?

#

If there’s a mismatch in CAN speeds, nothing works

fierce sage
coarse sun
#

Was just checking as it was a thought

fierce sage
coarse sun
#

Double press reset now

fierce sage
#

i did

after checking with those commands CAN falls into BUS_OFF state

image.png
coarse sun
#

You’re double pressing reset on the Katapult flashed board?

fierce sage
#

yes

coarse sun
#

I’m out of ideas here if your cabling is all checking out

#

<@&1304550334839918672> any other ideas?

fierce sage
#

Looking like this. I dont know if it helps.
Carthographer still disconnected.

20250402_174831.jpg
jovial rapids
#

Isnt that the resistor position?

#

Can data is in the sideways port?

coarse sun
#

I think that’s just CAN data there and not power

jovial rapids
#

Ah

#

u2c doc implies 120r can be placed there

coarse sun
#

Can we see your terminations at the other end?

jovial rapids
#

Looks like you could also simply run them both into the screw terminals for L/H

coarse sun
#

You can

jovial rapids
#

Im not sure that position can be used... BTT docs state it is connected to the MCU and has PIN definitions.

uneven minnow
#

use the two jst ports next to it

normal dune
#

picture shows a jumper on CAN pins too which I am pretty sure is not supposed to be there if it is not the end of the run

uneven minnow
#

correct

#

if you have an ebb and a afc lite out of those can wires, the jumper should not be in place on the u2c, but should be on the other two devices.

fierce sage
# normal dune picture shows a jumper on CAN pins too which I am pretty sure is not supposed to...

Just to clarify: i'm currently in a bit of a rush and i took the photo just to show you guys the position of the Can L and H.
I know the resistor shouldn't be there and normally isn't. I mentioned that the sb2209 also wont get recognized by my raspberry with katapult installed. I tried to disconnect the Boxturtle and see if i can get the sb2209 into a working configuration.

What you see is just U2C and Sb2209 on the bus. Both terminated.

Normally, there are only the resistor jumpers on the afc lite and sb2209 NOT on the U2C so please ignore it on this picture

fierce sage
jovial rapids
fierce sage
#

i still cant follow through entirely... even if i damaged one u2c on some way, the other one would still work. but it does not. how is it possible that the guides are that clear and after checking every possible issue i can imagine nothing is working. even the sb2209. i found a guide about one faulty katapult version causing that exact error. i tried to follow those steps aswell but it didn't fixed it for me like it did for the others.

ref: https://discord.com/channels/1229586267671629945/1343305254820773948

thorny oyster
#

@fierce sage For now, stay with a single U2C until you get UUIDs from all devices. Dual U2C needs special setup to work reliably (but once it works, it works well).
Make sure to physically unplug the second U2C and reboot the Pi.

#

Just never connect AFC-Lite and Carto at the same time.

#

Are you using Klipper or Kalico? I'm asking because Klipper's can bus query does not list devices once they have Klipper flashed onto them, while Kalico does.

thorny oyster
fierce sage
fierce sage
thorny oyster
thorny oyster
fierce sage
thorny oyster
#

Ah, ok.

#

Do you still have both U2Cs connected to the Pi?

fierce sage
#

no. just one as in the picture above. 2 of those were just a test but i stuck with all devices on one can network with carthographer disconnected.

#

no second u2c connected.

thorny oyster
#

Ok. That would have been too easy 😉

#

We can add the second one later, but only once everything works again with just one.

fierce sage
#

yea kinda giving me a headache sind i dont know were the issue could be anymore

thorny oyster
fierce sage
#

none of those devices showing a uuid while following the guide. doesnt matter if i use the first u2c or the replacement.

fierce sage
#

i just didnt remove the second one yet

thorny oyster
#

Even if you don't attach any CAN devices to the second one, it can cause issues if it's connected to the Pi.

fierce sage
thorny oyster
#

Ok. What does ls /sys/class/net/ on the Pi report?

fierce sage
#

i got a Voron Tridex with 2SB2209 on one U2c + Carthographer and some other machines using can.
All of them working fine. I just dont understand what the issue might be...

thorny oyster
#

That looks fine.

normal dune
#

I would start oover with the guide from step 1 just to verify nothing has been broken by accident

fierce sage
normal dune
#

I just meant Esoterical's CAN guide not a whole SD card and PI

thorny oyster
#

Can you temporarily connect the SB2209 to another printer with working CAN to see if it appears?

#

CAN is nice when it works, but diagnosing it when it doesn't is... not pleasant.

fierce sage
#

all my other machines were set up NOT using Esotericals guide and they worked... I dont know if they are outdated so i tried Esotericals Guide the first time with no success as you guys can tell... Everyone of you used this Guide aswell?

thorny oyster
#

Yeah, pretty much. It's the CAN bible nowadays.

fierce sage
fierce sage
normal dune
#

yeah Used it several times recently with no issues. If you flashed over USB for katapult did you maybe forget to remove the USB jumper on the SB229?

thorny oyster
fierce sage
fierce sage
thorny oyster
#

One more assumption to verify: do the red and black wires coming from the WAGOs carry 24 V and GND?

#

(hm, they probably are as the LEDs are flashing)

fierce sage
#

just to clear up my mind
checked:

  • wires (several times)
  • terminating resistors and their expected Ohm-Values (several times)
  • continuity (several times)
  • Pins
  • Voltage
  • Blinking lights (several times)
  • Buck, instead of USB
  • config (several times)
  • guide (several times)
fierce sage
normal dune
thorny oyster
fierce sage
#

after checking via python query the Network goes into BUS_OFF state and wont recover until reboot or manual reset. looks exactly like mentioned here: https://canbus.esoterical.online/troubleshooting/no_uuid.html

like if i force can to do something the network collapses.

image.png
Esoterical’s CANBus Guide
No UUID

A guide for setting up CANBus hardware on 3D printers

fierce sage
thorny oyster
#

Ok. That's good. So the BUS-OFF thing sounds like a problem.

fierce sage
fierce sage
thorny oyster
#

How sure are you that it's really L to L and H to H? BTT seems to assign pins randomly each time they bring a new device on the market.

#

There is no consistency between different toolhead boards or between toolhead boards and the U2C.

fierce sage
#

i'm sure since all of my u2c are version 2.1 with the same "fixed" firmware. the pins on which they are connected are in line with the screwteminals( continuity, 0 Ohms) i checked all pinouts and checked each C_H and C_L endpoint to endpoint via continuity

thorny oyster
#

Hm. That wouldn't detect swapping L and H on the toolhead board, would it?

#

That's something you could just try, btw. Swap H and L on one side and see what happens.

fierce sage
#

it would if the resistors are removed, wouldn't it?

thorny oyster
#

I'm not so sure 🤷‍♂️

#

I'm not sure if you already mentioned it above (it's late here 😉 ), but right now you only have the SB2209 connected, no other CAN device, right?

fierce sage
#

i checkt the cables if yellow leads to yellow and then plugged those into CAN-H and the same with CanL
so even if the color is mismatched the contacts should be correct...

fierce sage
thorny oyster
#

Ok. Just wanted to be sure.

fierce sage
#

doesn't matter i can switch around all possible configurations in no time if we can test something

thorny oyster
#

So in that case power off everything, swap L and H on the U2C side, power on and test again.

fierce sage
#

Only sb or sb and bt connected?

thorny oyster
#

Only with the SB. With only one device connected, it's easier to troubleshoot.
But before you do that, maybe take a photo of the SB2209 where we can see the CAN connector.

fierce sage
#

Only sb2209 with terminating resistor on sb and u2c

#

CanL and CanH swapped

#

sb2209 blinking

image.png
#

when i switch L and H around again it falls back to BUS_OFF after manual reset

image.png
normal dune
#

from the 2209 manual

thorny oyster
#

Ok. In that case switch it back.

fierce sage
#

So per continuity test, the upper one is connected to the can L on the U2c and same for CanH-CanH

thorny oyster
#

So if it's not the bus, and assuming the U2C functions, then it pretty much has to be Katapult.

#

Do you have the SB2209 with an RP2040 or with an STM32 chip?

fierce sage
#

STM32

thorny oyster
#

Can you post a screenshot of the Katapult config?

fierce sage
#

sure

thorny oyster
#

Damn. That looks correct.

fierce sage
#

i follow the guide bit by bit since it did not work the first time

#

since then its rather just messing around trying to identify the issue 😦

thorny oyster
#

Possibly stupid questions, but let's question everything:
You did compile Katapult for the AFC-Lite before going back to the SB2209, didn't you?

fierce sage
thorny oyster
# fierce sage yes

After you changed the config back for the SB2209, did you run make clean; make before flashing?

fierce sage
#

yes

thorny oyster
#

Damn. Why can't you just make a small mistake that I can find? 😉

fierce sage
#

trust me i'm hoping for the same thing

normal dune
#

might have to summon esoterical

thorny oyster
#

I'm honestly out of ideas right now.

fierce sage
#

he already had a look into it

thorny oyster
#

Except moving the SB2209 over to a known functioning printer.

fierce sage
#

gosh like frying the carthographer due to stupid circumstances isn't enough

normal dune
#

are you using the original U2C or the new one?

fierce sage
#

but i cant imagine that both of them are dead and if they are the 3rd will follow

normal dune
#

hmm, Did you try just the AFC-lite? maybe the SB is done

fierce sage
#

might be something software/pi related?

thorny oyster
#

Did you try any of the other connectors on the U2C (except for the USB connector; that one's special)?

normal dune
#

pi and U2C are on the same ground right?

thorny oyster
#

They should, because of the USB cable. But doesn't hurt verifying that.

fierce sage
thorny oyster
#

If that doesn't give us any clues, I have one idea left:
Reduce the baud rate of the CAN bus to 100k. You have to adjust that on both the network configuration in Linux as well on the device side (Katapult). They have to match perfectly.

#

If that also doesn't help, I personally would connect an oscilloscope and look at the waveforms. But that's me, and I own an oscilloscope.

normal dune
#

I will state that BTT for whatever reason does not include the Dupont in their CAN-OUT diagram so maybe something is weird there

thorny oyster
#

And one more: I think both Katapult and Klipper have scripts to query the CAN bus. Did you try both?

uneven minnow
#

I woudl suggest trying the jst xh ports first

#

start with one device, with termiantions on both u2c and that device. get that working. then swap to the other device, verify that works. then try to get both working.

normal dune
#

do you not need to supply the u2C with 24V power?

fierce sage
fierce sage
thorny oyster
#

CAN requires not only L and H, but shared ground between all devices, too. So it's really a 3-wire system.

#

Damn, why didn't I notice that.

fierce sage
uneven minnow
#

ah

thorny oyster
#

@fierce sage: before anything else, connect GND. It can't work without.

uneven minnow
#

then use screw terminals with same thing.

fierce sage
#

I have to schedule this for tomorrow. Its late here and i got an exam about electrical engineering this week (oh the irony)

Gonna test this and give you guys some feedback.
i really really appreciate your support!

thorny oyster
fierce sage
#

set all the devices to share the same GND level.
flashed a fresh version of katapult to the SB2209.
Issue persists

image.png
#

falling back into BUS_OFF state

image.png
thorny oyster
#

Damn.

#

I'm pretty much out of ideas.
I think you've already checked the cabling (conductivity from one end to the other).
You could try reducing the baud rate (down from 1M to 100k; both on host and in Katapult) to see if something causes issues with the high speeds.

#

But other than that I don't know where to look. If we could rule out either the U2C or the SB2209 as the problem, that might help.

#

What does dmesg say after the bus goes to state BUS_OFF?

fierce sage
#

reduced the speed on AFC and Firmware to 500k and still the same error.

image.png
#

ill try to start anew and use the old way to do can
like
auto can0
iface can0 can static
bitrate 500000
up ifconfig $IFACE txqueuelen 128

coarse sun
#

AFC-Lite won’t function at the lower bitrate btw. We ran into that once before

thorny oyster
#

Fascinating. Didn't hear about that.
@fierce sage in that case try 500k with the SB2209.

fierce sage
#

or any other speed 😦

coarse sun
#

SB2209 came back at a lower bitrate?

fierce sage
#

unable to check rn

fierce sage
fierce sage
#

Since i ran a last test before taking some time off, none of the ways worked out. I even went back to bullseye and did the manual configuration of can like it was working before. Still no solution in sight. at this point i'm guessing that all devices (or at least both endpoints) are dead even if they are blinking. there is traffic on the RX line but nearly none on TX. a new AFC is going to get shipped to me and i will try it with this board alone. if this works i'm pretty sure i fired my sb2209 aswell. I guess me and Can never gonna get along. i would like this issue to stay open until i return because if a solution is found in the end i want to share it so others have not to go through

uneven minnow
#

not a problem

fierce sage
#

So share one common GND ist prefered but seems not to be needed. My Tridex works flawlessly without u2c connected to GND.

<@&1304550334839918672> Further i received the AFC lite 1.1 and it showed up at first try.

I guess my boards got fried somehow.
Just a quick update. I gonna do some more testing when i return

thorny oyster
#

If the Pi and the devices on the bus are powered by the same PSU, for example, then the U2C shares that ground via the USB cable to the Pi.

#

Good news that you now have something that works 😀

fierce sage
thorny oyster
#

I wonder how the SB2209 got fried though.

normal dune
#

Hot plugging maybe?

fierce sage
forest coyote
#

Katapult seems to upload correctly but when I double click reset the board disappears from the list of by-ID devices. However the status led is flashing. If i hit reset again status led goes solid and I can see the board. However - I can't get it to show in katapult mode.

uneven minnow
#

Please open a dedicated help thread for your issue so we can track it

forest coyote
fierce sage
#

<@&1304550334839918672> So I just returned from my vacation and got some new boards. AFC, SB2209 and Carthographer got replaced and everything working fine now via two seperated Can networks. I guess i hotpluged something by mistake and the hole bus went to heaven.
I appreciate everthing you guys tried to help with but the issue was clearly on my side

thorny oyster
tropic wing