It’ll be something like flash a new
It’ll be something like: flash a new firmware to one side, see if the LED blinks, and vice-versa.
42 Replies
Here's two firmwares: one for each side. Load them both. When disconnected, one will blink and one will stay steady. When connected, if the connection is good between the two sides, they will both blink. Otherwise, one will blink and the other will stay static.
Backup your VIA settings if you like them because idk if they will get wiped.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
Just for completeness, do you mind snapping a video with those two firmwares when they’re pulled apart? Just to make sure they’re working the way I’d expect.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
Alright, thanks for testing. connection seems good. Mysterious.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
I can build a snap firmware that enables debug logging and we can see if there’s anything interesting there after flashing it onto both sides.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
I’ll get you something this evening 👍
This one has debug logging enabled. You'll want to clear the EEPROM on both sides first by following the "erasing firmware" section here: https://github.com/nullbitsco/docs/blob/main/bit-c-pro/user_guide_en.md#erasing-firmware
And then flashing this firmware on both sides. The debug messages can be read through QMK toolbox. You should see something like what's shown in the screenshot once it's connected.
The first few "failed to execute slave_matrix" are normal, since the slave side boots slower than the side connected to USB. However, those messages should stop once the other side fully boots.
@ahren_with_an_h not sure if replies to thread actually result in notifications so sorry for ping
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
Yeah, that’s exactly what’ll happen if the two sides can’t talk, which is how the board has been acting. It’ll do it forever. Alright, this is pretty insanely odd.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
Yeah, I should clarify a bit: that’s the message it prints when the USB side can’t talk to the other side. So you’d expect that if they’re not connected, but not if they’re pushed together or connected via TRRS.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
Yeah exactly. It knows there should be another half, and so it’s always looking.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
You know at this point, it’s probably not a bad idea to just try. The two sides are connected, but the protocol between them is pretty high-speed and sensitive. So if the connection is there, but not perfect, it could be causing issues. Perhaps try with just the KB_COMM pin on each side when you have the chance?
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
I would normally expect it to be connected or not, but I’ve also not seen a marginal connection on this pin before, so it’s possible this is exactly how it shows up.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
You’re welcome! Thanks for your patient debugging. This has been a strange one. We’ll get it figured out.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
You could try the pogo pins and the TRRS jacks too.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
That’s a good idea. If that STILL doesn’t change anything, we’ll go with the nuclear option. We’ll cross that bridge when we get there.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
Yeah, that is a good call and would be possible — connecting the two together via the communication pin along with a few other connections (power, ground, and a resistor to set the side on each) directly would be a great way to eliminate anything on the board itself.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
No worries, I could help you do that to test if you wanted to. Great idea on the alligator clip as well. That accomplishes almost the same exact thing.
You know one other thing you could test that’s purely firmware is flashing those test firmwares from earlier, but swap the sides. The one labeled left on the right etc. the reason is it uses one side as an output and one side as an input, so if half of the circuitry driving the pin is blown we would perhaps uncover it.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
Swapping MCUs is definitely a great next step. And a breadboard is always handy to have no matter what. I promise, we will get this board working again 🙌
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
Thanks for checking! I’m really interested in what’s up then. It’s mysterious because that test is more or less reproducing what the two sides are doing via the firmware (toggling a pin and reading it back on the other side), albeit much slower than it happens normally.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
Awesome! Well — good find. I’m glad it’s not the board. Really interesting that they “kind of” work. I’d expect a full failure, at least when it comes to a blown out pin.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
All great points. I’m happy help debug and if you do end up figuring out which is the problematic one (If it is just one, which seems totally possible) I’d be interested in getting it back just so I can do a failure analysis and make sure I know what went wrong.
Unknown User•2y ago
Message Not Public
Sign In & Join Server To View
Hey sorry, have been away/busy so I’m slow to reply. That’s great news! If you don’t mind, send it to:
nullbits attn: Jay
117 E Louisa St #126
Seattle, WA 98103
Feel free to use whatever is the cheapest/slowest shipping option that’s easiest for you. I’m thankful to have it back, no need to spend more than necessary to get it here! Thanks again. I really appreciate it.