The BBC have stopped by our offices at Tinker.it! to review a bunch of different Arduino projects. There was a great deal of gaffer taping, shuffling about, rigging lights, posing and sweating. The upside will be much better quality video footage of the Brockenspiel in action than there is already – and I didn’t even have to open up Final Cut Pro to do it. After being coaxed to explain what a microcontroller is in the simplest terms possible (and several takes to get it right), I think I could hardly put two words together anymore.
Last year at Dorkbot London I gave a demonstration of the Brockenspiel and fielded a few questions afterwards. Some clever guy in the audience asked if I could hook up a bar code reader – a thought I’d had but not implemented. When I said it would probaly work just fine, as long as the serial protocol is the same, someone in the audience called out “Hang on a minute, I’ve got a barcode reader around here!” (Only at Dorkbot!) Hidden in the depths of Limehouse Town Hall, he hauled out the reader, and by a stroke of luck, it worked. We were playing the music on beer bottles in a matter of minutes. Later, I made a quick video to demonstrate the concept.
Well I’ve finally got round to posting some video of the Brockenspiel doing its thing. I’ll be discussing how it works in more detail at Dorkbot, but you can get a little preview of the action below. If you can make it to the event, you’ll have a chance to try out your own swipecard and discover what song is encoded on it.
Well, the Brockenspiel’s been running for about 6 months now without any real hitches – 3xcept when somebody pulled the power supply out. The only thing not quite right is the mounting system for the tubes and a couple of slaphappy solenoids. Got to get round to posting that video… Looks like I’ll be demoing the brockenspiel at Dorkbot London in April.
The song player, now dubbed the “Brockenspiel” has been installed here at the lab. It’s nice to see it up and running. Still have a few things to work out. In particular, the Arduino board seems to fall over after a few days. The first time this happened I had to remove the board, rewrite the software and re-install it. This fixed the problem. Still need to determine whether this is a coding issue or a hardware design flaw.
Most of the day was spent mounting and testing the rest of the solenoids. These little gems put out a nice 5 grams or so, depending on how much the spring is loaded and how much current they get hold of. They will do fine with a 100ms duty cycle.
I ganged them up on the back of the magnetone frame and used hot glue to adhere them to cable ties and to keep the cables managed. The glue also came in handy for wrangling the return spring. It was not attached to the body of the solenoid, so I had to come up with a clever way to do so – the hot glue was just the trick. I don’t know how many cycles it will last, but this stuff is pretty sticky, so I am pretty comfortable with it in the short run. A longer term solution with some sort of mechanical fixture would be better. As soon as I get hold of a laser cutter…
Last, but not least a soft rubber tip is needed to strike the chimes. Metal to metal is harsh, so I’m dulling the plunger tip with…drum roll…a little squidge of hot glue! It’s the salve of the prototyping gods…
The last little to-do, was to begin the breadboard for the circuit. I borrowed todbot’s clever idea to make a homespun Arduino Shield out of a perfboard and some pin-strips. In the end, the two will mate perfectly and the daughterboard will be easily removable for any maintenance or debugging…
The circuit diagram for the magnetone is pretty straightforward. The cool thing about the ULN2003A is that it has current-limiting resistors built in, so you don’t have to add those as separate components for each of the devices that you’ll be triggering. Any suggestions for improvement are welcomed!
I’m calling my current project the “Magnetone” because it uses a bank of solenoids to play a set of chimes. The interesting part is the data source. At the Lab we all have magnetic swipe keycards for access to various areas. Now nobody knows what’s encoded on these cards because the information is not exactly easy to read without the right tech – and anyway who wants to bother with reading them? I wanted to reveal that data in an interesting way. We all carry around little “possible melodies” with us and the magnetone can manifest them physically.
The project uses the Arduino prototyping board to send a trigger to a ULN2003A darlington array. This allows the proper current to be activated to several solenoids which strike the set of chimes. The data source will be a magentic swipe card reader which sends regular ascii keystrokes via RS-232 to the Arduino.
Comments (0) 
Themocracy
