Jesse from the Utah Ballroom Dance Company approached Oregon State University's Electrical Engineering Department to have some wirelessly controlled, light up dance suits. The interns at TekBots were given the task of researching and developing a system that can wirelessly control 6 suits with 8 channels of Electroluminescent Wire each.
We quickly distributed tasks to each of the interns (Tuan Truong, Chelsea Collette, and Brian Benavidez). The system logically broke down into three sections: the transmitter, the receiver, and the driver, the latter two being the devices physically attached to each dancer. Chelsea took on the driver, Brian took on the receiver, and Tuan took on the transmitter. Here is rev 0 of the boards:
The first boards worked for the most part but there was clearly a new revision in our future. The boards were larger than necessary (check out all that blank green space!) and would cost too much to produce them in quantity. The receiver needed a lot of additional work, extra hardware and unused microcontroller pinouts needed to go, and the power distribution needed to be addressed. This board could take in 12v which would be passed directly to the inverters needed to drive the wire at 100V, 2kHz, then regulate the 12v down to 3.3v to power the microcontroller. This was easy enough but we were wasting power dropping 12v to 3.3v using a linear voltage regulator. The solution here was to use a less power-hungry switching regulator to drop the 12v down to 5v, then use the linear regulator for the additional drop from 5v to 3.3v. The benefit here was that we could add a couple pinouts from the microcontroller to drive addressable LED strips like these:
Now we could offer Utah Dance the option of using LEDs in addition to EL wire. Awesome!
So we went to work on the second revision, adding some hefty power hardware and combining the receiver and driver into one, more compact board. The result:
Here is the board with all of the wires connected and ready to go (yeah, it's a lot). So we 3D printed a case and crimped on some JST connectors to try and clean everything up.
That pretty much sums up the progress so far. Obviously there were many more small decisions made and hardware/software development which will come to light in future posts, but for the sake of length, we'll leave it at that for now.
No comments:
Post a Comment