Modular RGB Chessboard

Status: on hold
Progress: 20%

Build Log

FYI

  • this project I did not document the building process so I am going back much later to build this log.

  • this in its self is V3 of my attempts at making a chessboard type of project. The other 2 attempts i don't really have and want to go back and document so just know the first 2 didn't end up working correctly due to hardware issues in the PCB's.

The idea

  • This Project is a modular 8X8 grid of squares with a magnetic sensor to detect pieces being moved, I designed a PCB that would talk i2c back to a Raspberry Pi 4 running the chess engine and the communications. each PCB had 4 LEDS on them to light up so it can provide visual clues to people who dont really know how to play.

  • Unforantly V3 teachinaly works, its just a major pain to reload the programming on the microcontroler thats on the PCB. This was too annoying for me and i lost my interest in making this project. Also I spent wayyyyy too much money on parts that did not end up working how I wanted it, so I moved on to new and more exciting project

  • here we have the render of the PCB I made
    the PCB

Blockers

  • I managed to get pretty far, I spent many hours hand assembling and soldering all the componets on each PCB, in V4 (if i get inspired again) I will defiantly find a better way to do this, I will most likely have the PCB fab assemble them.

  • the coding part is not my strongest, I like the engineering more, i find it easier and more enjoyable to do this than trace down bugs. when building this AI was not a big thing yet, going forward I will use it to improve my coding.

  • the sensors them self would also give false readings ever now and then that would throw off the code, maybe that could be solved in code or maybe hardware.

Lessons

  • The project it self might have been a failure, but I dont look at it that way. This was one of the first few projects where I would design my own PCB, I taught my self an entire new field and managed to get a functional product at the end. It was far from perfect but I learned the hard way, how I should better design my PCB's for my next projects.

  • around this time I also started to look into other CAD programs, at the time I was using TinkerCAD. This web based CAD software is fantastic, it was simple enough for anyone to understand but also powerful and allowed me to do what I needed. However, my skills and projects have reached a complexly that they could not scale with, so I was looking into Fusion360.