suziesparkle - I'm building a Tidbit and I've s...
I'm building a Tidbit and I've socketed the MCU in order to test what I have so far. I am new to this and frankly I don't trust my soldering! When I short each switch's pads I get the output on the blue paper. Should I be getting multiple keystrokes like this? I was expecting single keystrokes at this point.
The rotary encoder is working for volume so I'm super happy about that at least 👍The kit is really nice and I'm going to use it to speed up my photo processing workflow.
Solution:Jump to solution
Thank you again for looking at this. After a very drawn out investigation I realised that all my cheap MCUs all have a fault with pins GP26-29 being connected to earth (I am not alone it seems https://www.reddit.com/r/olkb/comments/1fm1o32/rp2040_cant_use_gp2628/). Coincidentally, these are the column pins on the Tidbit. This problem never showed up on my first project because it didn't use those pins. On the plus side, I now know more about keyboard circuits than I ever wanted to! I can make work by removing the header pins on the bad GPIOs and wiring the columns to unused pins. It will not be pretty but it will work.
It's academic now, but to answer your questions: The socket is Mill-Max 115-47-624-41-001000 and pins 3128-1-00-15-00-00-08-0. The LEDs work perfectly, which I am very pleased with because they are the first thing I ever surface-mounted 🙂
Thanks again for taking the time to help....
Thanks again for taking the time to help....
7 Replies
Unknown User•3w ago
Message Not Public
Sign In & Join Server To View
That is correct. Do you know where would be a good place to look for the short based on the results? I don't see any bridging so I'm just not sure where to start!
Unknown User•3w ago
Message Not Public
Sign In & Join Server To View
Thank you for the offer! I have added photos here: https://www.flickr.com/gp/suziesparkle/2oK36EehPe I honestly don't think there's anything to see but an extra pair of eyes is always helpful.
Given how it works, the most likely cause would be a row shorted to ground, but I went over the PCB with a multi-meter and could find nothing connected to ground that shouldn't be. Today I swapped out the MCU for a spare and exact the same problem, so it seems unlikely it's that. I've also now compiled my own firmware now so I can start isolating each row and col, and see if I can find anything that way. Even if I can't fix it, I think I can still make something useful with it. Frustrated though 😦
Unknown User•2w ago
Message Not Public
Sign In & Join Server To View
Solution
Thank you again for looking at this. After a very drawn out investigation I realised that all my cheap MCUs all have a fault with pins GP26-29 being connected to earth (I am not alone it seems https://www.reddit.com/r/olkb/comments/1fm1o32/rp2040_cant_use_gp2628/). Coincidentally, these are the column pins on the Tidbit. This problem never showed up on my first project because it didn't use those pins. On the plus side, I now know more about keyboard circuits than I ever wanted to! I can make work by removing the header pins on the bad GPIOs and wiring the columns to unused pins. It will not be pretty but it will work.
It's academic now, but to answer your questions: The socket is Mill-Max 115-47-624-41-001000 and pins 3128-1-00-15-00-00-08-0. The LEDs work perfectly, which I am very pleased with because they are the first thing I ever surface-mounted 🙂
Thanks again for taking the time to help.
Thanks again for taking the time to help.
Unknown User•2w ago
Message Not Public
Sign In & Join Server To View