In my final semester of mechanical engineering (undergrad), and taking an Electronic Musical Instrument Design course (yes, that is a class!). Being a mechanical engineer, I may be getting way in over my head:
My dream for the final project is to create a spaced out grid of sensors (copper tape) and use CapSense to simply trigger outputs based on touch. Each sensor (or row) would represent a sound. I would also like to have a "setting" that sweeps the columns of the grid, reading individual rows over time, creating music based on the grid squares that are triggered by metal blocks sitting on them.
The outputs would be, hopefully, midi messages using simple if/then statements (i.e. if cap is triggered, send midi). There are music maker shields that have built in midi sound libraries, that I believe can be simply transformed into audio by the shield itself.
After much research, things I still don't understand:
1) I would like to have a fairly large grid (128, or 8x16 would be great). Most of my research so far is in i2c expanders. Is this the best option? Can CapSense work with port expanders? It seems CapSense is merely timing the difference between signal sent and received, so what complications could result by reading inputs from a large amount of I/Os?
2) For those who understand music maker shields: can midi messages be sent from arduino into the shield, and how does this communication actually take place between the shield and arduino? Does the shield have the capability to transform the midi into audio, like I assume it does? Is this idea of triggering midi signals to make sound even possible?
Extra for those willing to tackle it: If the music maker shield can write to the sd card, how difficult would it be to record midi signals over time and loop them on command?
Please let me know if I am out of my mind, especially if the complexity of this is insane. One thing I do have going for me is a large desire to learn and a decent amount of free time...
Thanks for your help!