Completed! Touija Board!

This project was definitely an exercise in design and engineering.  I had originally created an XY axis motion rig on a frame just to test the electronics:

That worked pretty well. I used an Arduino Uno, an Adafruit motor shield v1, nema 17 motors, motor mounts, some 8mm rods, linear bearings, and timing belts.  I made the platforms out of plexiglass because I had some lying around.  I used the Adafruit Stepper motor library, which also uses the AccelStepper library.

Then it came time to design the table itself. I began with the table top, since I had a good idea of the size I wanted. I figured I could reverse engineer it from there.

Then came the hard part.  Retrofit engineering the XY linear motion rig to fit underneath the table top.  I ended up completely starting over because the first prototype was designed to be sitting on a table or floor and the eventual magnet platform would be mounted to the top carriage pointing upwards.  For the tabletop, I intended to mount the hardware underneath the plank, and the magnet carriage would need to be pointing towards the top of the table, which is still up when it’s right side up, but is down when the table is upside down for install.

Once that was sorted, I just had to make the platform that the magnets would sit on.  I don’t have any pictures of that process, but it was mostly trial and error.  Then it was time to test the planchette movement!

In this video I was moving it with my hand.  I was having code issues.  I eventually worked those out, basing my code off of the MultiStepper example from the Adafruit Stepper Motor library.  I was able to get x and y movement, but I discovered that without a way to zero the planchette — to make sure the system knows where the planchette is — it was impossible to try to plot out where each letter was. I added in two momentary push buttons in line with the carriage and the magnet platform so that I could set a basis for movement.

Once that was done, it was possible to actually start plotting out letters.  I did a test video at this point, because I knew that I’d be replacing the vinyl letters and that to plot out where they were now would be pointless.

 

 

Two cans of spray paint and some new vinyl cuts later…. I had this:

IMG_20141214_163226

Lookin Spiffy!

Then it was a matter of plotting out the letters, and getting it to receive the tweets.  I used this tutorial by Lucidtronix as my jumping off point. I have a processing sketch that pulls the latest tweet of whatever search term I specify from twitter, and sends it over a serial connection to the Arduino, which then passes it to a function that cleans it, and parses out each character, telling the planchette where to move.

Overall I am happy with this build, but I will continue to work on it to make it as I had originally envisioned.  The motor shield I’m using does not leave many pins for other functions, so I’ll be replacing it with the newer motor shield from Adafruit.  This will enable me to add a wireless breakout board that will take the place of the current wired USB serial connection.  Once that is done, I will translate the processing code to a Ruby script that will run on my linux server that is handling many of the other functions for my thesis.  Then I can build the rest of the table and its protective plexiglass vitrine.