Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Building your own EEG machine

  1. Aug 6, 2010 #1
    As useful as an EEG machine would be for me, I can't afford to buy one but I'm going to build one in the near future. There is a lot of information on this site
    and from what I've read so far, building an EEG machine is no simple matter. Not for the layman at least. I've never even used a soldering iron before, would you say I'd better practice building simpler circuits and devices before attempting to build an EEG machine?
  2. jcsd
  3. Aug 6, 2010 #2
    AFAIK the only "challenging" part of an (non-invasive) EEG is the input amplifier that gets the signal from the electrode and makes it "loud enough" to do something with. The problem is noise and interference rejection. But I doubt that it's much worse than your typical microphone pre-amp. In fact if you can find an audio preamp that works down (very close) to DC you could probably use it our-of-the-box.

    If you've done no electronic projects before, start with a few kits and toys, just to get the hang of it. http://www.goldmine-elec.com/ has some learning kits, and might even have a mic pre-amp in the list.

    When connecting things to living tissue you need to be _very_ careful about grounding and the possibility of electric shock, so start out with devices that are powered by low voltage batteries and don't ever break the skin.
  4. Aug 6, 2010 #3


    User Avatar
    Science Advisor
    Gold Member

    My advice would be forget about making a EEG machine.

    Have you read the http://openeeg.sourceforge.net/doc/WARNING.html" [Broken] on their site?
    Last edited by a moderator: May 4, 2017
  5. Aug 6, 2010 #4
    Question; Why do you need a EEG?

    I just cant see the usefulnes of a homemade EEG, even if you get some good readings, you still need to know how to analyse them. Using the readings to other things than check for sleeping disorder requires advanced learning algorithms and lots of data.

    Also what dlgoff said. Serious health risk involved!

    Build a normal sound amplifier instead is my advice.
  6. Aug 9, 2010 #5


    User Avatar
    Science Advisor

    I remember reading about a guy who had designed a wearable EEG for his dog (complete with cute bone-shaped enclosure worn on the dog's collar) once in Circuit Cellar. It logged the EEG data and detected how often, and when, his dog was having seizures.
    http://www.cc-webshop.com/214-May-2008Electronic-Data-Logging-And-Analysis-SA-2008-214-41.htm?categoryId=-1 [Broken]

    Unfortunately, that article isn't available unless you buy it or have a subscription. Still, I echo some of the other sentiments in this thread about the utility of your own EEG (unless you're a doctor or medical researcher, or some such)
    Last edited by a moderator: May 4, 2017
  7. Aug 10, 2010 #6
    Will do thanks. I haven't even grasped all the theory yet, I got up to transistors but got sick of reading all the theory without putting any of it into practice. I suppose its about time I started.

    Yes I read the warning. I'm too determined to be swayed by that though.

    I want to see what brainwave patterns are predominant when I'm in different states of mind i.e. sleep paralysis, on psychoactives, hypnotic trance. What would I do with a sound amplifier? Are you experienced with reading these things? I don't have too much time to put into learning this so maybe I'll this isn't a realistic plan I'm starting here.

    I'm planning on studying neuroscience once I get my chemistry BSc so it would definitely come in useful in the long run but maybe I'll take this project off my list of priorities for now.
    Last edited by a moderator: May 4, 2017
  8. Aug 10, 2010 #7
    Good on powering through the scary warnings and attempts to discourage exploration of horribly too-complicated-for-amateurs topics. I wish you luck, and hope that you don't electrocute your dog or whomever is your experimental subject...

    My advice on electronics is: Just skip the transistors, ignore (almost) all of the theory, and build some kits using Integrated Circuits. Then you'll know if you're cut out for the field.
  9. Aug 10, 2010 #8


    User Avatar
    Science Advisor
    Gold Member

    As long as you read it. And being hard headed can be a good thing. :smile:

    I'm not experienced with the readings but there was a time years ago that I was involved with biofeedback and used one. The first thing that I had to learn before being ready to use "feedback" from the EEG machine was to simply control the temperature of my hands. A simple temperature sensor on my finger with, in this case, filtered pink noise as the feedback. It didn't take long to change the my hands temperature 15 deg F. Now the eeg machine I was hooked up to had at least a dozen channels IIRC. And again, filtered pink noise was the feedback mechanism. With some practice, you can put yourself in a sleep state while being couscous, albeit very very relaxed. This particular eeg machine was designed with all the feedback hardware built in. It was an interesting experience and certainly helped me to realize how to eliminate stress.
  10. Aug 10, 2010 #9
    I can't brag about much experience in the field, but from a signal analysis aspect I can say getting cohering data will get tricky. Signal patterns will differ alot depending on many factors. And if you are not able to tell the different from a propper reading and a faulty one, your likely too get a wrong perception about the subject, which will do you no good in neuroscience.

    If you want too learn, read a book. No point in reinventing the wheel.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook