Octopus 1.1 connection issue

[Arcane3D]

Printer Model

Voron 2.4 350mm, Formbot Kit

Extruder Type

Clockwork 2

Cooling Type

Stealthburner

The first Octopus 1.1 that I received had a bad USB-C Port, I was never able to connect to it with the RPI, and I validated by using a voltage tester. No draw at that port with the jumpers properly configured.

I ordered a replacement and gave BTT the benifit of the doubt that the initial board could have just been damaged by me somehow. I know how to handle boards, so I am very doubtful that this happened.

Second board worked perfectly fine, firmware copied over, USB-C connection to Pi was flawless.

After getting the "preflight" completed and finally 1 successful full print, I decided to install the back plexiglass. As I was doing this, I flipped the power switch and killed power to the 2.4. Upon reboot, I have the "Klippy host software is attempting to connect"... error and my log indicates that the pi can't access '/dev/serial/by-id/*...' at a previously working serial. I have rebooted many times in hopes that a firmware reset would fix this.

Is there a way to reset this issue, or is this board fried as well?

 

[ah2049]

I'll assume you you've checked to make sure all the necessary cables are still attached, so does the octopus shows any signs of life, such as lights coming on?

If it lights up, have you hopped on the Pi to see if it sees it by using lsusb?

Possibly try a different USB cable?

Last thing I'd try is to reflash firmware on the octopus. Perhaps something got corrupted.

 

[NoGuru]

Also try and just restarting the firmware if you can get connected to the web interface. I have to do this sometimes on a machine for it to connect properly.

 

[angrygriffin]

Might you have fallen foul of the Debian bug?

Debian Bullseye Bug causing Klipper to no longer find the printer board

Moved to Debian 11 (Bullseye) udev bug - No board or serial found

klipper.discourse.group

 

[ChrisNeal]

Might you have fallen foul of the Debian bug?

Debian Bullseye Bug causing Klipper to no longer find the printer board

Moved to Debian 11 (Bullseye) udev bug - No board or serial found

klipper.discourse.group

Thanks for posting this. Last night I flashed my Octopus twice and both times the firmware.bin file got renamed to .cur, but the serial/by-id command wouldn't produce. All this was after being frustrated with USB-C not powerring my PI 4B and trying 3 different cables, then finally getting results from a USB-A (Octopus) to C (PI).

 

[Sanity Agathion]

Ahhh yep the infamous bug. Glad you worked around it.

 

[ChrisNeal]

Thanks for posting this. Last night I flashed my Octopus twice and both times the firmware.bin file got renamed to .cur, but the serial/by-id command wouldn't produce. All this was after being frustrated with USB-C not powerring my PI 4B and trying 3 different cables, then finally getting results from a USB-A (Octopus) to C (PI).

And... I thought I'd figured out a workaround, a way to use a different USB port. Nope. All I was getting from the Octopus USB-A was power for the PI. I was excited for a few minutes because I'm new to all this and "Wow... there's the Mainsail logo!" Talking to the OS through my notebook was great for a few minutes, but soon I realized there was no communicating with the MCU. I spent most of the day trying different pins and plugs and flashing and reinstalling, eventually giving up on the original board and installing a second one I have.

 

[Sanity Agathion]

And... I thought I'd figured out a workaround, a way to use a different USB port. Nope. All I was getting from the Octopus USB-A was power for the PI. I was excited for a few minutes because I'm new to all this and "Wow... there's the Mainsail logo!" Talking to the OS through my notebook was great for a few minutes, but soon I realized there was no communicating with the MCU. I spent most of the day trying different pins and plugs and flashing and reinstalling, eventually giving up on the original board and installing a second one I have.

You connected RPi to USB-A on the Octopus? Where did you connect the data connection from Pi to Octopus then? You may be missing something.

Normally traditionally you can take power either from separate 5V PSU for RPi, and connect USB-A on RPi to USB-C on Octopus. Or you take power from Octopus to power Pi (there is a special cable for that in kits usually), but you still need to have some communication link between Pi and Octopus. Either USB-C or UART.

 

[ChrisNeal]

You connected RPi to USB-A on the Octopus? Where did you connect the data connection from Pi to Octopus then? You may be missing something.

Normally traditionally you can take power either from separate 5V PSU for RPi, and connect USB-A on RPi to USB-C on Octopus. Or you take power from Octopus to power Pi (there is a special cable for that in kits usually), but you still need to have some communication link between Pi and Octopus. Either USB-C or UART.

Sanity: I'd read in a few places when I was planning this build that I could forego the 5V PSU. So maybe in saving a few measely dollars I went and cost myself some headaches. Originally I tried USB-C to USB-C, but the Pi was not getting any power. So I tried USB-C (Octopus) to USB-A (Pi), also with no power. Then I pulled power to the Pi from the USB A (Octopus) to C (Pi) and thought I could get data through that cable as well. As you mentioned that doesn't work, so I tried using 2 USB cables at once.. 1 for data and 1 for power. That didn't work, so I had to start reading about the Pi connector and GPIO pins, then crimping and trying combinations of those, none of which ever got my Pi talking to my Octopus Pro 1.0.1 F429.

So I swapped out the MCU, used USB-C (O) to USB-A (Pi) + used the UART2 pins to provide power to the Pi (see picture). It's ugly, but I have better connectors on the way to make a permanent all-UART to GPIO cable . I'd like to remove the USB cable from the equation because of bulk and seeming easy to damage the USB-C port on the Octopus.

<edit to add> And the first board.. I'm not sure what might be the deal with it. It never got dropped or anything and it's humid as heck right now, so static is a really low risk and care was taken anyway. So it's sitting in a box taking a time-out

 

[Sanity Agathion]

What kind of Pi do you have? Octopus board may not provide enough power to power RPi 4 and you ,ay get undervoltage warings.

 

[ChrisNeal]

What kind of Pi do you have? Octopus board may not provide enough power to power RPi 4 and you ,ay get undervoltage warings.

I have a Pi4B with 4gb. I've also read what you're saying about the Pi potentially not getting enough juice. My thought was maybe the 350W PSU that I installed might clean up the power enough to make for a reliable 5V.

Just a few minutes ago I tried building a firmware.bin to use PD5/PD6 pins for data, and I flashed the MCU, then trried connecting those pins to GPIO and removing the USB cable. All I got after restating was a note about "Klippy trying to connect..." So I went back to using the USB cable for data and jumper wires for power, and reflashing the board to use those. The new board works again.

So there's a number of complications that're my doing here: 1) I have very low experience with Pi 2) I went and left out a 5V PSU, which I should order now 3) There was so much swapping around of cables, jumpers, fw's yesterday that yeah I probably could be missing something. 4) Also with all the Octopus board versions available, it was a little confusing at first.

 

[Dan_The_Printer_Man]

I have a Pi4B with 4gb. I've also read what you're saying about the Pi potentially not getting enough juice. My thought was maybe the 350W PSU that I installed might clean up the power enough to make for a reliable 5V.

Just a few minutes ago I tried building a firmware.bin to use PD5/PD6 pins for data, and I flashed the MCU, then trried connecting those pins to GPIO and removing the USB cable. All I got after restating was a note about "Klippy trying to connect..." So I went back to using the USB cable for data and jumper wires for power, and reflashing the board to use those. The new board works again.nwell

So there's a number of complications that're my doing here: 1) I have very low experienceanwe me with Pi 2) I went and left out a 5V PSU, which I should order now 3) There was so much swapping around of cables, jumpers, fw's yesterday that yeah I probably could be missing something. 4) Also with all the Octopus board versions available, it was a little confusing at first.

Yeah meanwell 5v psu would have been the way to go.

 

[Sanity Agathion]

You can maybe also buy a reliable buck converter to power your Pi from 24V. ALthough 5V PSU will have some additional advantages.

 

[bruce356]

Sanity: I'd read in a few places when I was planning this build that I could forego the 5V PSU. So maybe in saving a few measely dollars I went and cost myself some headaches. Originally I tried USB-C to USB-C, but the Pi was not getting any power. So I tried USB-C (Octopus) to USB-A (Pi), also with no power. Then I pulled power to the Pi from the USB A (Octopus) to C (Pi) and thought I could get data through that cable as well. As you mentioned that doesn't work, so I tried using 2 USB cables at once.. 1 for data and 1 for power. That didn't work, so I had to start reading about the Pi connector and GPIO pins, then crimping and trying combinations of those, none of which ever got my Pi talking to my Octopus Pro 1.0.1 F429.

So I swapped out the MCU, used USB-C (O) to USB-A (Pi) + used the UART2 pins to provide power to the Pi (see picture). It's ugly, but I have better connectors on the way to make a permanent all-UART to GPIO cable . I'd like to remove the USB cable from the equation because of bulk and seeming easy to damage the USB-C port on the Octopus.

<edit to add> And the first board.. I'm not sure what might be the deal with it. It never got dropped or anything and it's humid as heck right now, so static is a really low risk and care was taken anyway. So it's sitting in a box taking a time-out

image is not worth posting or viewing