mcu 'mcu': Unable to connect?

[ken226]

My Voron 2.4 has been working great since I built it.  This morning I turned it on and saw that I had some Klipper and OS System updates available, so I updated everything and rebooted.

 

Now I'm getting the "mcu 'mcu': Unable to connect" error.   

 

My first thought was that the OS update had changed something in the PI's USB settings, so I swapped in another (one of three backups) Pi's that I have preconfigured to run my printer.

The backup Pi also couldn't connect to the MCU.  I swapped in and tried all of my backup Pi's and none could detect the MCU.  On all three Pi's, the  command "ls/dev/serial/by-id" returns "No such file or directory".  When I run lsusb,  only the webcam shows up.

I thought, maybe somehow the Octopus V1.1 board lost its firmware, so I went through the Voron docs instructions, remade the firmware.bin file, and using the SD card method I re-flashed the Octopus board.  The firmware re-flash seemed to work, as after the flash the filename on the sd card had changed from firmware.bin to firmware.cur.   But, after reflashing the firmware, still, none of the pi's are able to detect the Octopus baord.

The board is powered up, and all of the LED's are on, but it just isn't being detected.  I even tried connecting the board to my Windows PC with a USB cable.  The PC doesn't detect it either.  I tried other USB-c cables, same result.

 

Is there anything else I can do, or check on the Octopus board before I order a replacement.

 

The Pi still works fine, the USB webcam works in all of the USB ports, and with three different Pi's,  so it definitely seems to be a problem with the Octopus board.

 

 

Edited May 1, 2023 by ken226

[mvdveer]

Just had a read on discord and found this:

 

[mvdveer]

Here is a temporary solution. Posted it under a new topic as not to hijack your thread.

[ken226]

8 minutes ago, mvdveer said:

Just had a read on discord and found this:

May be part of the problem if refreshing the MCU does not work:

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1035094

Yea, I already tried reflashing the mcu, and it made no difference.  

I'll read through that link and try the fix suggested, but,  earlier I swapped in two other Pi's. A Raspberry Pi CM4 and a BTT CB1, neither had not been updated. Neither of those were able to detect the octopus board either.  Nor would a Windows 11 PC.

[ken226]

22 minutes ago, mvdveer said:

Here is a temporary solution. Posted it under a new topic as not to hijack your thread.

I went to that thread and installed the Debian working version backports.  It didn't make any difference.

 

I've reflashed the MCU, it didn't work. 

I also tried two other pre-configured pi's that havn't been updated in a few months, and it didn't work either.  Once was a CB1 that I used with this printer before I managed to buy a Raspberry Pi, and the other was an Orange PI 4 Lts, also a unit i ran this printer on before I started experimenting with the CB1.  Both of those were configured and working with this printer before. 

 

Also,  if that Debian update messed up the USB ports, wouldn't have effected the USB webcam? It still works on all 3 of the Pi's ive been using to diagnose this.

[mvdveer]

2 minutes ago, ken226 said:

if that Debian update messed up the USB ports, wouldn't have effected the USB webcam?

It did not affect the USB ports physically, but the symbolic link to the USB ports. Thus klipper does not know it is connected to the Octopus board through USB.

If you have your pi connected through UART, this should not apply.

Will do a bit more research on this.

[mvdveer]

When you turn your octopus boar on, is the red LED under the mcu lit up? You should have three LED's on when powering the board.

[ken226]

4 minutes ago, mvdveer said:

When you turn your octopus boar on, is the red LED under the mcu lit up? You should have three LED's on when powering the board.

Four total are illuminated.

[mvdveer]

That means you have power to the mcu

 

[ken226]

It seems like the USB port on the octopus died.   Three Pi's and a windows PC couldn't detect it, and it's definitely powered.

[mvdveer]

4 minutes ago, ken226 said:

It seems like the USB port on the octopus died.

May be it, I take it you tried a different USB-C cable between the octopus and the Pi as well?

[ken226]

9 minutes ago, mvdveer said:

May be it, I take it you tried a different USB-C cable between the octopus and the Pi as well?

 

Yes, I tried two cables.    Then to be sure, I double checked both cables by connecting my phone to my laptop with them.  They both transmitted and received data just fine between my motog power and Lenovo laptop.

 

It's just so weird, this update and the octopus USB port dying, at the exact same moment.   Maybe, just an improbable coincidence.

[mvdveer]

17 minutes ago, ken226 said:

 Maybe, just an improbable coincidence.

It is strange indeed and I am not one to believe in coincidences. Will keep researching and if I find anything else will let you know. But you have done all and more trouble shooting and it seems it may be the USB port. You can always measure the current in the port to confirm. If you can be bothered, this is a good link on how to do this.

I have one of these, but the above method may help to diagnose a faulty port.

[ken226]

6 minutes ago, mvdveer said:

It is strange indeed and I am not one to believe in coincidences. Will keep researching and if I find anything else will let you know. But you have done all and more trouble shooting and it seems it may be the USB port. You can always measure the current in the port to confirm. If you can be bothered, this is a good link on how to do this.

I have one of these, but the above method may help to diagnose a faulty port.

I'll see if I can find a spare usb c connector tomorrow and test the port.

 

I have a new Octopus board on the way from Amazon.  

[ken226]

12 hours ago, mvdveer said:

It is strange indeed and I am not one to believe in coincidences. Will keep researching and if I find anything else will let you know. But you have done all and more trouble shooting and it seems it may be the USB port. You can always measure the current in the port to confirm. If you can be bothered, this is a good link on how to do this.

I have one of these, but the above method may help to diagnose a faulty port.

I checked the Voltage output of the Octopus usb-c port with my Innova 3340a and there is nothing.   If Plugged into the Raspberry Pi, 5V is present, but that is expected because the Pi provides 5v through the USB ports red wire.   Current draw is 0 between  the Pi and USB-C measuring with the multimeter in series with the red wire.

 

[bdejong11129]

Just had the same thing happen to my new build.  Was running as it should until I updated.  Then it will not see the MCU and if I use Putty and browse to the /Dev/Serial/By-Id there is no By-ID folder.   There is only a By-Path folder now.   Doing the above command and using By-Path does not seem to fix it.  

This is very frustrating since I had just printed the first test square. 

 

 

[Jerazmus]

I had the same thing happen to me. I actually had to rebuild on the pi imager and re get a new firmware for the mcu. I tried reflashing  with the firmware on my pi that was existing but to no avail. Once I did that and reflashed the newly compiled fw and pi, then it came back. It was the strangest thing. 

[DennisM]

Just stumbled upon this thread -- sounds quite scary honestly and makes me hesistant about updates, although I know they're important.

Any particular update that you know of that might have triggered this? Klipper or something else? Or do you run "update all?"

 

Thanks!

[TomTom2005]

have same problem for a weeks now, try all the same stuff, new cabel, reset all things. Do i not a OS update after the new install, all works fine.