For my final project for Physical Computing, I attempted to create a device that would allow the user to control a sculpture via brain waves. The device is comprised of a hacked Mindflex (a toy EEG) that,via a processing sketch and an Arduino microcontroller, controls a variable magnetic field, based upon the user’s level of concentration. This magnetic field, in turn, alters the shape of a highly magnetizable fluid (ferrofluid).
Ferrofluid in a Magnetic Field:
Ferrofluid is a liquid suspension of iron nanoparticles that are small enough to remain suspended by Brownian motion (meaning, the iron does not separate easily from the suspension liquid). This liquid becomes magnetized in the presence of a magnetic field and a cone like pattern emerges in the fluid along the lines of the applied magnetic field:
As seen in this experiment performed by scientists at MIT, when confined to two dimensions and subjected to a variable magnetic field, ferrofluid takes on lesser known, otherworldly forms.
Process (What Not to Do):
After reading fellow ITP student Eric Mika’s extremely helpful blog entry on his Mindflex hack, I cracked open the case to the Mindflex headset, loaded the Arduino brain library, connected my Arduino, and began reading data packets from the serial monitor. The Mindflex toy works by sending the EEG readout from the headset to a receiver on the base, which in turn controls a variable DC voltage that regulates the speed of a fan. I thought that if I could locate the variable DC voltage, I would be able to somehow connect it to an electromagnet to create a variable magnetic field. So, I cracked open the Mindflex base, which required removing many hidden screws and breaking plastic pieces.
I considered creating an electromagnet from scratch by winding magnetic (copper) wire around a soft iron core made of coat hanger pieces.
I had considerable difficulty working with ferrofluid. Essentially, ferrofluid is both extremely messy and extremely sensitive to magnetic fields.
The combination is disastrous. Ferrofluid will stain any container that you place it in (glass or acrylic) and find even the smallest of cracks in said container to leap through, when a magnetic field is near. Placing the ferrofluid in an isopropyl alcohol/deionized water mixture is supposed to reduce the staining power of the ferrofluid.
However, in my experience, the mixture (even in varrying concentrations) alters the consistency of the ferrofluid (transforming it into a chunky metal suspension in water). Ferrofluid is also very expensive and not so widely available. Thinking that my first ferrofluid purchase might have been of inferior quality, I ordered another bottle from a different company.
Unfortunately, the staining/consistency issues where repeated with the second batch. Finding an airtight container was also an issue. I sealed a petri dish with a hot glue gun in an attempt to replicate the 2D surface of ferrofluid used in the MIT experiments. I am still cleaning the resulting mess from this experiment.
Weary of the effect of the strength of the electromagnet on such a unstable substance/contraption, I decided to switch to using a magnetic field created by servomotors and rare earth magnets. The EEG readout from the headset (via a processing sketch and an arduino) controls the motion of a servo motor, which has a rare earth magnet attached to it.
As it stands now, I have created a machine that succeeds in making a massive mess. Over break, I hope to create some form of my original idea. I’m hoping to find a way to fashion a leak proof container of ferrofluid made of a material that does not stain.