Today was another big day for the guitar hero bot. Firstly, we finished up wiring all of the internals for the guitar, including wrapping the wires in heatshrink tubing to keep them tidy. Additionally, we cut an opening at the top of the guitar to act as a receptacle for the 9V battery that powers the propeller processor running in the base of the guitar. We cleaned up the edges of the battery opening, and it actually looks pretty good now, considering there is a 9V battery there. Wires from the battery run down the neck of the guitar to where the processor and board will be situated. We considered using the USB to power the processor, but didn’t want to risk limiting the current flow to the guitar or to the processor.
Our goal is to have the guitar both look normal when complete, and to also be usable in a manual mode. For the connection from the guitar to the sensor bar, we are equipping both the bar and the guitar with 15-pin D-sub connections. 5 of the pins will carry photodiode signals from the sensor bar to the guitar, 1 will be reserved for an extra sensor if we need it in the future, and 2 will run power and ground at 3.3V from the microprocessor to the sensor circuits (there are unused pins). A standard cable between these connections can simply be unplugged if desired. A switch will also be employed allowing the user to deactivate autonomous mode so the user can play manually if desired.
We also completed our first working draft of our programming today, and HAD THE GUITAR PLAY A SONG ON EASY. It worked! Or well, at least it did until the power went out half way through the song!!! : ( Talk about bad timing. But before the power went out, it had gotten about 70% of the way through “slow ride” without missing a note. I’ll film it working tomorrow and be sure to put it on youtube (hopefully before the power goes out again). Anyways, the program now delays appropriately, and is able to filter out noise and strum on both single notes and chords. And don’t think this came easy. Hours of time was spent debugging this code to get it to work, and I’m sure lots of tweaking still needs to be done. But now, we have to worry about working out bugs and getting circuit boards printed so we can make it look real pretty :).
Here are some pics from today for you to enjoy.
