the project

I am building a MIDI-controlled air horn organ.

Plans:

• controlled by a Raspberry Pi A+
• MIDI-based with possibility of human input as an extension
• 64-note range
  

Progress so far:

• MIDI files can be processed using python in order to track what notes should be sounding at a given time (data point at each note-on or -off event)
• processing made possible by python-midi
• control of a single shift register (mostly) functional
• I'm using SN74HC595PWR latching shift registers from Digikey for signal-level output
• control is based on py-spidev
• power control tested, theoretically functional
• ULN2803ADWR 8-channel driver for power output, supposedly can take 500 mA per pin
• miswired something and fried one of the shift registers and the RPi2 that I was using to test
• determined that there are no HDMI cables in the house to hook up the RPiA+ to the TV
• determined that my WiFi adapter won't work with Raspbian (OS on the Pi)
• got functional WiFi adapter
• learned to solder
• designed and ordered first iteration of the controller PCB from oshpark
• tested version 1 control modules (3x)
• successful test, see video
• eliminated one possibility for solenoids as non-viable
• refined and ordered final version of control PCB from oshpark
• noticed that I forgot to correct silkscreen when I moved one of the headers
• found another source of solenoid valves at a relatively reasonable price
• designed an optoisolated interface board
• safety!
• ish
• assembled final version of control PCBs
• determined that the aforementioned solenoid valves have insufficient airflow
• designed pneumatically-operated gate valve
• laser cut design
• cheap (if it works)
• later proved non-functional
• realized that I shouldn't use 28 gauge wire for running the common positive for the solenoids
• decided that it would probably be good to just shell out the $15 or so for the 9-position connectors to rig for connecting solenoids to the control boards
• acquired a regulator
• reconstructed CAD files for gate valve
• fully parametric
• fancy input form to edit dimensions
• post linked again here
• preliminary testing of gate valve completed
• keyboard layout selected
• chromatic button accordion layout
• keyboard panel laid out and cut
• wore out two hole saws
• they're dead, Jim
• failure to chooch
• keyboard panel assembled
• don't ask me to change it
• I won't do it.
• Spoiler: I did it.
• input PCBs designed and built
• I messed up the first time
• they should work now
• keyboard rebuilt
• new panel using 1/2" MDF
• cut on the ShopBot
• properly-spaced buttons so it's easier to assemble
• redesigned valve concept after testing showed it non-functional in its planned configuration
• later scrapped this design as well in preference for larger pre-made solenoid valves
• redesigned interface PCBs to make use of easier connectors (RJ12 and IDC) and surface-mount components 
• also changed input PCBs to reduce wiring harness requirements for the keyboard
• started manifold design
• new valves linked above now thread directly into tapped holes in low-pressure supply pipes
• need to choose between round valve clusters or rows in line

Things to do:

• lay out final horn arrangement
• build lots of air horns
• tune lots of air horns 
• tune even more air horns
• check complete control system
• devise user interface (aside from keyboard)
• final construction