Shitty Amateur Electronics - Part 7: Finishing The Matrix Board

The last time we saw the matrix board it had a row of LEDs, some resistors, and a single row of header pins. After realising that doing the matrix soldering wasn’t as hard as I thought it’d be I basically threw away any thoughts of putting the board to one side during the Edinburgh Fringe in August and ended up just cracking on with it.

I have a couple of photos here of the board in various stages of completion, 3 rows and 5 and a bit rows. There’s also a couple of shots of the back of the board, showing soldering with varying degrees of success - but every connection worked fine when tested with a multimeter.

Adding the final row of header pins on the other side of the board finished the component part of the build.

Which meant it was on to the wires. I was worried about this bit. I didn’t have any good wire stripping tools and I figured the process of measuring and cutting would be difficult. I was right, but only partially. The third arm gadget really helps with this, but ultimately having bits of wire just resting on the board in the right position isn’t really working. The best way to do it may be some Blu Tack on the wire to stick it down first. The other helpful bit with this process was the fact that I got wire that maintained its shape when bent, so I could make some runs that were neat. I maintained that tidiness for precisely three wires.


Looking more closely, you can see the burn marks, the crappy solder joints, and the close proximity of some of the joints. I really worried about short circuits here, but continuity tests came out fine on the meter, so... There's also the definite problem with the extra stripped insulation. I've taken off way too much, but I got better at it as the job went on. Clearly for more intricate designs I won't have this much leeway, and with stripboard it becomes even more important as you may inadvertently bridge something together.


And so here it is. One finished board ready for interfacing with the Raspberry Pi. There is the small matter of the code to sort out, though... I'm not looking forward to that, but there's a few examples I can find out there that may help. One thing I'm wondering is if there's a library for interfacing with this kind of matrix. There are for ones with controller chips, but not for ones where you're hooking up direct to GPIO pins as I am. I may get a proof of concept working and then see if I can make a nice wrapper for people. Define your rows and columns, the pins in use, and then have a Python class to represent it.


Of course, there's another way I could go with this, and that's to use an Arduino or similar microcontroller kit. This has certain plus points, but some negatives as well. Harder to set up, but less resources and similar afterwards. The board is technically compatible with anything that can send signals to the individual wires, basically. I could even make a second tiny board as a controller, use a shift register, etc. But ultimately the RPi setup is the way to make this easy while I'm learning.

Unfortunately before I even started the code for this I started making a new board. More in the next post...

Comments