how to make (almost) anything

Output Devices: LED Arrays

Week 10 Tools/Resources:
Eagle
Roland SRM-20 Mill
Soldering Equipment

Charlieplexing tutorial
Make: AVR Programming
Fab Acad: Julie Garnier

Charlieplexing

With the final project in mind, and building on input devices, this week I worked on learning the basics of designing and controlling LED arrays. Although I understood how Charlieplexing should work at a diagrammatic level, once I began to dig into understanding the way Neil's board was set up, I found it difficult to grasp the specifics. Visualizing the communication between the various pins and LEDs helped me see how everything was connected--and the way things intersect at each individual LED-- more clearly.

Pin communication on Neil's board

Board Design & Fabrication

I took Neil's board as a starting point, but arranged a smaller number of LEDs linearly instead of as a grid. There was an idea I ran across in the book Make: AVR Programing, which creates low res images from a series of linear lights in motion utilizing similar techniques as other persistence-of-vision examples. When I was partially through soldering the components to my board, I noticed that I had failed to make the connection from the 2x3 header to the ATTiny PB3 pin. I decided to continue fabricating and to make the missing connection with a wire; a technique I have yet to try.

Bridge to nowhere...

A decision that seemed like a great idea during design, but ended up being a huge pain, was my choice to separate the light strip from the rest of the board, connecting the traces via wires. I thought this would give me more freedom of movement for awesome visual creations, but I didn't leave a lot of real estate to ACTUALLY SOLDER THE WIRES TO THE BOARD. This led to some of the copper peeling away from the backing, which I was able to alleviate in some cases by soldering all the way along the trace to the next component. On the last wire on the light strip, the copper did pull all the way off. As a last ditch effort, I bypassed the orginal trace completely, bridging from the more generous border of the board to the LED with a 0 Ohm resistor. That should work, right? Even though this was a fabrication full of unforced errors, it was good to have enough experience at this point to be able to think of my feet and troubleshoot on my own, in the moment. We'll see how it goes.

Sloppy this week.

Programming

Not a great start.