I mean the prototype sensor node for my house CAN-bus, decked out in MSU colors. So with the exception of possibly the next node, no other circuit boards will have a theme or use primitive construction techniques. The real sensor node will use a PIC18F2580 SOIC package, have a board less than 50cm square and use the shield concept swiped from an Arduino. At least that’s what I put together to have made by someone else. As a side note, I’d like to know what perf board would have hole spacing of √(.02)” to allow the screw terminals shown jammed in diagonally.
Since this 40-pin microcontroller is under-utilized, I’ll use it to test more nodes in the future. It was quicker to spend a day or so cutting, stripping and soldering wires than wait weeks for real circuit boards. This will at least allow me to start a test bus, even if it only has a couple nodes.
The next steps are software, and how to use the built in features of CAN smartly. A lot of this is already in spec form, like SAE-J1939 for heavy machinery that costs money all the way to OpenCAN that’s less money. These schemes are more for systems that are time critical and people’s live being at risk. None of these are applicable to a normal house, so a new scheme can be developed. My goal is to release open source boards and basic code libraries for hackers and to provide design services for people who don’t think any of this is fun (weirdos).