Ask HN: Help me create my own feelSpace belt

5 points by adk 17 years ago | 3 comments

First, some background on what it is I am looking for:

http://feelspace.cogsci.uni-osnabrueck.de/en/index.html

http://www.wired.com/wired/archive/15.04/esp.html

Basically, it is a belt with a bunch of vibrating pads going all around and an electronic compass. Whichever pad is pointing north, vibrates. The end result is that "the tactile perception of the vibrators [disappears] whilst conscious information processing is displaced by a new 'sense of direction'."

I have wanted one for a long time. At first I hoped that modular components would be easy to find and put together. Apparently not so. I then toyed with the idea of making it out of old nokia vibrating batteries and an electronic compass component, but I quickly realized this is way out of my area of expertise. So then I thought, no problem, I'll hire someone with the right skillset. Except, I don't know what the skillset is, or where to start looking.

I figured there would be someone in the HN community who'd be able to point me in the right direction.

sasmith 17 years ago |

I built a mediocre prototype around November, and am currently working on a hopefully better prototype.

Useful info about the following: I am a cheapskate, and am interested in producing this very cheaply. If you're willing to drop $80 on the compass module, certain aspects of this will change.

My first prototype was built out of cell phone vibrating motors, a belt I stole from a friend that had good holes for the motors, and a compass from http://www.robsonco.com/Dinsmore/. It worked alright, but the motors were prone to getting stuck, the compass module wasn't as consistent as I'd like (eg, it gave around 30 degrees to cardinal directions and 60 degrees to diagonals). To build this, you'd probably want basic soldering skills and preferably a basic understanding of electronics. If you were seriously interested I could send you instructions.

For my second prototype I'm using coin-type pager motors, so I don't need to worry about the heads hitting anything. Specifically, I'm using: http://www.pagermotors.com/index_files/Products/DiskL/DiskL.... (and fwiw I measured them at 40mA). I'm also using an AK8973 compass chip, although I haven't yet tested it. This is _not_ a hobbyist friendly chip, and it's kind of a pain to get. On the other hand, I got them for $7 a piece and I expect it to give me better accuracy than the Dinsmore module. I expect to use a microcontroller interpret the output and activate the motors (I may need more cir. For the belt I'm essentially using a canvas strap. To do all of this (mostly to deal with the newer compass chip) I've so far needed to etch my own circuit boards (you can get around this pretty easily for approx $20), will need to solder better than Santa's elves, and will need to program a microcontroller.

adk 17 years ago | |

sasmith, thanks for the information. Would you send me your email address? (andrej.kostresevic@gmail.com)

jurjenhaitsma 17 years ago |

I read about this a while ago (slashdot?) and thought it was quite intriguing myself. I believe there were some complications with withdrawal symptoms so buyer beware.

As for your request, most competent electronics hobbyists / engineers could likely build a crude version, but I could see complications arising from the vast number of electrical sources potentially disrupting your "north" signal.

Which led me to ponder a refinement - by using gps, although encoding location is quite a complication, it would be an incredible extension of the idea. A possible implementation would have latitude and longitude pads that could either vibrate at a varying frequency or (for extra leet points) activating a series of binary pads to encode the lat/long.