The next project I’m going to be looking into is the RGB lighting for the hamster cage project (post about the cage; post about the intended design). I’m now back at the usual first stage of my projects: experimentation. This is usually the time where I just hook things up, see how things work, and generally mess around. To give you a rough idea, my experimentation now will be getting the ESP32 to connect to the WiFi; then, I’ll try and get an LED to blink; lastly, I’ll try and get the LED to blink depending on whether I’ve sent a command over WiFi. As you can see, it’s step-by-step increasing the complexity, until I am where I need to be.
The next step, usually, is design. Once I’ve tested out all the individual bits of my final project, and I have a rough idea of how to do it, it’s the time to bring it all together. My first step is always to make a schematic in KiCad. As usual, before I start I try to learn a little bit more and apply at least one new thing before I get started. That’s why I was so happy that I found this video:
I’m only about half an hour in so far, and I’ve already learned a tremendous amount about net labels and sectioning off components, as well as just some handy KiCad tricks. Just for that alone it was useful to me, and I still have two more hours of the video to go right now!
Aside from that, I wanted to find a good solution to the mess of wires I had in the Pomodoro design. The outputs for the ATtiny84 were all so close to each other, and some had to move upwards while others moved down – it ended up as a large bowl of spaghetti (a tangled, disordered mess of wires or connections are often informally called spaghetti, for obvious reasons). Yet, there are two clear bundles of data in the Pomodoro timer: the two sets of LEDs. So, I looked up how to do bus wires, leading me to this post. It’s an explanation of how to apply bus wires in KiCad, and using wire labels to connect everything together. That way, you can get a clear schematic, and KiCad can still read it and build a rat’s nest out of it.
When I learned these things, I just had to stop the video and apply this right away. The only schematic I had lying around was the pomodoro one, so I figured to just test out the process:
Now compare the old schematic to this new one that has all the lessons applied:
What a difference in clarity! No more crossed wires, and several subsystems are clearly separated out to their own space. I greatly prefer this schematic. The only downside I found was that the labels are later on also used in the PCBNew program to label the actual traces. Now, that may be quite a useful feature later on, but it also causes a little confusion. Take a look, for instance, at pin 7 (PA6): it’s both MOSI as well as the Pause/Reset switch. Now, since I applied the MOSI label to it to connect it to the ISP, the whole trace was called “MOSI”. However, I’d much rather apply that label only to the trace leading to the ISP, and Pause/Reset only to the trace leading to the button. Maybe that’ll be possible, maybe it won’t. More to learn!