Let's pull the time machine out of the closet and crank the dial way back to a much simpler time. Yes, today we are traveling back to 2004. You may wonder why we are going to 2004? The answer is quite simple. In 2004, I built, and for the first time satisfactorily documented, one of my strange electronic art projects. As you may have guessed this project was called "Trigger".
"Trigger" is built inside one of the infamous Hanging Chad voting machines (that threw into disarray the 2000 presidential elections). It was modified to read-out right or left wing propaganda, depending on user interaction. Towards what side of the political spectrum it swayed, and how extreme it became, depended upon three factors; the direction that the case was rotated after it was picked up, the severity of the degree to which it was turned and also the amount of time in which it was rotated to that position. In this way, users could pick up the case and choose their own adventure through the political spectrum.
In the fall of 2004, "Trigger" was on display at Parsons School of Design as part of the "Voting Booth Project." It was surrounded by artwork submitted by much more reputable artists like David Byrne, Christo, Frank Gehry, Milton Glaser, John Maeda, and Diane von Furstenberg. To this day, not a single one of them will return my calls.
"Trigger" was made in collaboration with Raoul Rickenberg.
Step 1: Go Get Stuff
This was documented in 2004. Some knowledge was lost to time, but here is a basic outline of components that you will need:
- A Hanging Chad voting booth from the 2000 presidential election.
- A Dinsmore 1490 digital compass
- A 2 axis accelerometer
- SPST switch
- Small portable radio
- 20mhz resonator
- SPDT switch
- A cheap radio
- An iPod
- An iPod remote
- Some circuit boards, components, batteries and stuff...
Step 2: Compass
Get a Dinsmore 1490 digital compass. Realize you can't stick into a breadboard. Solder a bunch of wires to it and get it working with some LEDs in four cardinal directions.
If you happen to be in Manhattan in 2004, like I was, you can find North by determining which direction is uptown.
When you are done, solder the compass and LEDs to a PCB. The LEDs are important because it will be near impossible to troubleshoot later without them.
Step 3: Accelerometer
Once you know which direction is which, finding left and right should be easy, right? WRONG!
The reason this is not easy is because left and right is dependent upon your orientation when you pick up the case. In other words, walk up to an object and pick it up. Note which side is left and which side is right. Put it down. Turn 180 degrees and pick it up again. Notice that left and right has changed.
So, how did I determine which way was left and which was right? If you read the step heading you would know... acceleration. If you knew in which direction the case accelerated when it was picked up, you could guess 9 times out of 10 which way the person was facing. From there you could use the compass to assign values for left and right.
Thing I learned about accelerometers:
- they don't work well on breadboard/prototype boards
- they like a long "leash" as you have a tendency to toss them around while prototyping (even if you don't need to)
- They can be a pain in the neck sometimes.
Step 4: IPod Remote
My collaborator was really insistent upon using an iPod. I felt there were better ways to solve the problem of audio playback, but I guess he just really liked the sound of the word. Remember, this is 2004 and people were still getting shot over their iPods. They were killer. So, everyone say it with me... eyeeeee-poddddd.... Doesn't that sound good?
Anyhow, the fastest, cheapest and easiest way to control iPods at the time was to hack a wireless iPod remote and use a PIC chip to hit the buttons (via 3 relays - forward, backwards, pause/play).
So, I opened the remote up and added two wires to each of the buttons and put the case mostly back together.
Step 5: Intermission
Now that you have those 3 elements working, go back to your West Village apartment and spend all night sitting on your futon and writing a lot of code to get them all to play nice with one another.
Step 6: SPDT Relay
Add an SPDT relay to switch between two audio inputs. One audio source is direct from the iPod and the other source is a "static machine."
Remember, this is 2004 and audio storage and playback is not yet a solved problem. However, radio noise is a solved problem. A "static machine" is basically a $2 radio radio that is tuned to a frequency with no signal. It makes a noise that sounds like "SSHHHHHHHHHHH....."
Step 7: Put It on Cardboard
Wire everything together and line it up on a long piece of cardboard. This should emulate it being inside the case and will make it easier to test directionality.
Step 8: Insert the Chip
Transfer the chip from the development board to the circuit board that you have been building. Expect nothing to work right, as not to be surprised when nothing does. Spend many hours troubleshooting until you are back to where you started from. Don't stop now! There is still some more work to be done!
A note on the program:
I don't have the code for this project any longer and couldn't even begin to tell you what it looked like other than it being a heckuva-lot of MBasic. However, to give a brief summary, the code calculated the initial relative position of the case when someone picked it up and then calculated the relative direction it moved from the initial position. All of this data then triggered the case to flip through dozens of audio recordings of political speeches on the iPod. When the case was put down, the iPod would be re-centered.
Step 9: Put It in the Case
Build out the inside of the case with Styrofoam. Put all of the electronics neatly inside and discover that it mysteriously stops working. Insulate everything with electrical tape to keep it from coming into contact with static electricity on the Styrofoam. Duct tape the "static machine" and 9V batteries hastily to the side of the case.
One last detail! Install a SPST momentary switch into the bottom of the case. This will determine whether or not the case is being held or sitting on the ground (I was initially planning on using an FSR, but the switch worked better).
Step 10: Close the Case
Close the case and run it over to the gallery with 15 minutes to spare.