Daily Archive for February 14th, 2012

Power LED Matrix for ACRIS

I wanted to demo the ACRIS boards at MIT’s annual HKN Project Expo. This is a simple expo run in part by the EECS department at MIT. The idea is to provide a space for EECS students to show off cool stuff they’ve made. There were all sorts of projects: software, vehicles (Jordan showed off Segboard!), tesla coils, and, of course, blindingly-high-power LEDs. 🙂

My goal was to show off the ACRIS boards because I one day want to turn them into kits. But, the only completed lights I had were the wall sconces I made. So, I decided at the last minute to build a big LED matrix using a piece of sheet metal a friend found for me.

How last minute? I started at 2am the day of the expo and finished at about 10am. 8 hours. 20 LEDs (80 pads to solder to including 20 triple-point-to-point connections), 80 wires. Yeah. I’m tired.

But the result was this really derpy LED matrix held together with lots of Gorilla Tape.

It all began with the top of a rack server (at least that’s what I think it was).

I divided it up so that I would have a 5×4 array of LEDs. My original plan was to drill holes in the board and then screw the LEDs down like I usually do. Unfortunately, MITERS was closed by 3am, so 🙁

Next, I set to work modifying my ATX power supply. I added barrel plugs onto some wire and hooked those up to the 5V line on my power supply.

I had already built the rest of my LED controllers. I have a little army of them now. 🙂

Oh crap. Is that the sunrise?

Time to get serious. I started soldering the LEDs together by connecting all of the VCCs together. My plan was to have 4 columns of LEDs with 5 LEDs per column. Each column would be controlled by its own ACRIS LED controller.

Soldering the VCC lines were tricky because it was effectively a triple-point-to-point weld. And I was starting to get frantic.

Next, it was time to solder more wires on. All these wires I needed to connect to the LED board too… Eep what a mess!

All right, that looks a little better. Gorilla Tape to the rescue!

Lots of soldering, continuity testing, etc. later, I wired everything up. By this point, it was like 10:30 and I had literally a few minutes to run over to the expo.

Close call! But it worked beautifully on the first try. By some miracle, even despite the sleep deprivation, I didn’t screw up any of the LED connections. I had written a really quick visualization algorithm to show the power of the LED board. I wanted to add more plugins, but didn’t get a chance to.

I ended up getting second place and winning a little money. ACRIS is a little more funded now. 🙂

I’m really hoping to get back to working on the boards again soon. I have a lot of new ideas I want to test out. My friend Scott is working on a USB-based version.