Node 2
Node 2 is based on the aggregate, semi-coordinated movement seen in factories, farms, and other mechanised environments. This device uses one of my favorite of the simple machines, the lever. The images and video are better descriptors of the node in action. The following is for those looking for a bit more detailed explanation of what is happening in the work.
This node is an array of eight steel arms; each of which is balanced on a sharp pointed fulcrum. Each arm has a 12 volt fan (the type used for cooling computer cases) at one end, and a reed switch at the other. This assembly acts as a rocker arm which is put into motion when as the fan turns on. Depending on the orientation of the fan, the arm swings up or down. When the fan is switched off, gravity does its thing, and the arm swings in the other direction. By varying the length of time each fan is on and off, the rate of the swinging can be (loosely) controlled. As the arm swings, the end with the reed switch passes in front of a stack of small neodymium magnets. The magnets cause the reed switch to close a circuit, which is registered via USB to the controlling computer.
The controlling computer is an old IBM Thinkpad running PureData. The PD patch is connected to an Arduino board via the Pduino object. The patch sends and receives protocol2891 commands, and uses them to control the pulses of the fans. It also reads the state of each reed switch, and sends it out to the wider world via the 'net. See the node2 tech page for a description of these commands, and what they do.
There is an accompanying PD patch for this node specifically designed act as an abstraction which can be used to generate sounds, visuals or any other PD triggered event from what is going on with this node. The advantage of using this patch is that all the settings are pre-configured. All you need to do is plug in your stuff to it, and away you go.
Essentially, the node functions as eight metronomes which can be controlled remotely. It was originally installed as a test at the PD Spring School 2007 to see how easy it was for PD beginners to interact with. The video is of that installation. Since then, I've been working on a handful of patches which generate sound and visuals based on the movement of the arms. As these are created, I'll post them here as examples that people can use / learn from.
Overall view of the installation
This was the first installation of the node, so I wasn't too concerned with fine tuning the layout. I arranged the nodes in a row mostly to easily see the differences in the way each device moved, and to help with diagnostics.
Detail of pivot
Each fulcrum consists of two bits of sharply pointed steel nestled into divots in the rod below the point. This gives minimal resistance when the arm is moving (important, as the fans aren't that powerful). It also adds a little visual tension to the device.
Detail of Fan
The fans are typical 80mm computer case fans. I first used these in my work a couple of years ago in "In the Bucket". The mechanism of which was originally inspired by the work of Andrew Saiia. I've kept with them for a couple of reasons. I try to use commodity items whenever possible in my work. I believe that the use of materials that are commonly seen in other contexts lower barriers for the audience to approach the work. In the instance of of using case fans, you might not recognise the object - unless you have looked inside a computer - but there is a good chance you will recognise the sound. The use of commodity materials also allows others interested in developing work along similar lines to use similar materials as the original work. I like to think of it as getting as "open source" as possible with the physicality of the device.
Detail of Sensor
The sensors used are small reed switches held in place with "crocodile clips". Opposite each switch is a stack of neodymium magnets. When the switch passes in front of the magnet, it closes a circuit (it turns on). As more magnets are stacked, the magnetic field becomes stronger, allowing the switch to be further away from the magnets and still be activated. Other sensor systems could also be used here. Light sensors, tilt switches, etc would all work. Magnets though, seem to be the least susceptible to outside interference, which could be considered a good or bad thing.
From below the fans
This shows the arrangement of the fans, arms and cabling from the rear of the device. The stands are designed to give a little "spring". This isn't noticeable unless the arms move quickly, or if a longer arm is attached to the device.
Side view
Here you can see the different length arms used on the individual devices. Using different length arm changes the dynamics of the assembly. Longer arms are less responsive to the force generated by the fans, resulting in a slower more sedate movement. Once it gets going, it tends to keep swinging for a longer period of time as well. Shorter arms change state (from movement to stillness, or the other way around) much quicker.
The Arduino boards
Two Arduino boards were used for this installation. One Arduino controlled the fans, the other read the sensor data. Arduino doesn't have enough inputs and outputs to control 8 fans and read 8 sensors. There are plenty of solutions to this problem, most of them more efficient and elegant than using two board. However, for this setup I was more interested in the interaction of devices with PD than I was the setup of the electronic hardware. As I had a couple of boards in the studio, this was the quickest solution. The solution is a bit of a kludge, but it worked. The next installation will likely have some sort of multiplexer which will allow me to use one board.
Click the image to play video.
(Opens in new window & requires flash)
This video shows the node in operation. The sound is audio generated by a quick PD patch I cobbled together to demonstrate how the node could be used to generate audio events.
Something you should know:
This project is part of the research supporting Ben Dembroski's PhD candidacy at the Glasgow School of Art. As such, any information generated during the course of correspondence or collaboration regarding this work may wind up in my PhD thesis for all the world to read. Please read the fine print and research methodology before collaborating with this project for full details what information may be disclosed, and the manner it will be presented.