Keyboard

I tested all of the switches and diodes with a multimeter to make sure they still work after desoldering them. There were a couple broken diodes. I’m going to rebuild this keyboard before I finish redesigning it because I never was able to find a flaw in the design with the original. I’m going to use the same parts as I removed so I just wanted to make sure they still work. I tested the switches using the continuity setting which was annoying because I had a hard time holding down the key and also touching both pins with the probes of the multimeter (no alligator clips). The diodes were much easier to test. I just set it to the diode setting and put the black probe to the side with the black stripe and the red probe on the other side. I looked for a reading of 0.55L-0.65L and then reversed the probes to make sure it was 0L.

I layed out keys, placed diodes and LEDs, and made some changes to the schematic. I’m not sure if I used the right footprint for the capacitors (especially the two bigger ones - I feel like they’re way too big for a keyboard), which is why I haven’t been working on this project much lately. This is how it looks in KiCAD right now. Nothing is routed yet, and as you can see in the bottom left, two of the capacitors are ridiculously big. I made the layout in keyboardlayouteditor and then used the position relative to tool in KiCAD to get the keys layed out properly. I still need to test the diodes and switches from after I disassembled the first version to know if I need to order new ones.

I made a new keyboard schematic. I’m going to redesign this keyboard to have more keys and use per key lighting. I’m eliminating the OLED, but using the same microcontroller. I have capacitors every 4 LEDs and input 5V halfway through the chain so there shouldn’t be any fading towards the end of the chain. I’m still not sure what the issue with the old design was but I hope to figure it out.

I completely built, programmed, and have now taken apart the keyboard. It took me a couple of hours of soldering to get all of the diodes and keys placed. There is a lot of time I spend programming that didn’t get recorded because I thought it would be quick. It was not. In the end, I used Claude and Gemini and was finally able to get the QMK compiler to work. However, the .uf2 didn’t make the keyboard work properly. I tried several different ones and used the serial monitor and send outputs to Gemini but wasn’t able to get it to work. The best I could get was random characters. So, I decided to spend a couple hours removing all of the components so I can try it again on a fresh PCB. I’m not sure what the issue is and I inspected the PCB so hopefully I drag soldered something and didn’t see it or installed a diode the wrong way. I have ordered a new microcontroller because the old one broke while I was trying to remove it. Also, apparently I didn’t take almost any photos.

I started assembly. I still don’t have enough funds for the PCB (Blueprint please fulfill my order for a grant from the shop 🙏). I inserted all of the switches into the 3D printed top plate , which is a necessary step prior to soldering the switches to the PCB.

CAD work - after the first print of the top plate I realized a couple of issues. First, I hadn’t made holes for the stabilizers. Second, and this was just cosmetic, there were partial labels for the diodes from the KiCAD .step export in my top plate, which made the edges of the key holes look messy. I fixed both issues.