Binaural Beat Hardware

[rich decibels]

Hi everyone, I'm trying to design a binaural beat generator for consciousness-enhancing brain wave modification. No, seriously.

For the uninitiated, the theory: when you are alert, the signals in your brain have an average frequency of 14-30Hz. A relaxed brain is slower, ticking along about 8-14Hz. Just before you fall asleep you enter the theta (hypnagogic) state, associated with meditation, hypnosis, hallucination, etc. with your brain waves averaging 4-8Hz. Below that is deep sleep and finally death.

The theory is that if you could subject the brain to external sources of these low frequencies, you could change the state of consciousness. The problem is that 8 Hz is well below the range of human hearing, so unless you have a room-sized sub-woofer, it is a little tricky. One common method of generating these Extremely Low Frequencies (ELF) is with binaural beats (BB).

With BB, one audio signal is sent to each ear; the frequency of one signal being a few hertz faster than the other. So with one ear hearing 500Hz, and the other hearing 508Hz, there's an 8Hz pulsing beat that's heard in the brain, but not in the ears.

So far so good. There's plenty of software around that will let you generate your own BB (search for bwgen or sbagen), and there are plenty of mp3's and videos around online. The thing is, I want to do this in hardware, with blinky lights, and gratuitous shiny bits and meters.

Any suggestions on how to precisely control the offset of two oscillators with a range of 0-30Hz would be greatly appreciated. I have a few ideas but it's always helpful to have a few extra minds weighing in on a design like this. The first thing that springs to mind is to use a high-accuracy ganged pot to control two 40106 oscillators, with a second pot attached to one oscillator to create the offset.

The thing I'm completely clueless of is how to measure and display the frequencies - I feel it is quite important to be accurate with this. Any thoughts on this would be helpful because I imagine it may alter the design approach from the bottom up.

Cheers,
Rich

[Top Top]

The problem I think you would run into is that as you change the frequencies, the difference between them does not change in an equal manner - ie: the beats between the two are not constant/fixed even if the two oscillators change mathematically by the same amount.

For example, two oscillators tuned to a certain interval apart will create a certain difference tone or "beat" against eachother. If you increase both oscillators by one octave, the "beat" I believe will also be a different frequency (no longer the same as it was at the first position). I'm not saying it's impossible, but the type of control you need is probably not going to be found in a lunetta-style package. Lunettas tend to be a little more crude when it comes to precise frequency control, and based more around the digital/logic tricks of the cmos chips.

The other thing is that squarewaves (which is what you will get from cmos chips) in such a concentrated/droning fashion will not help a person get into a relaxed state as they can be quite agitating when raw... you would probably want to filter it.

I've messed a bit with this, by the way, looking up frequency combos that are supposed to work (using my nord modular). I had poor results from listening purely to drones, but found that out of everything, sine waves in mid-low frequencies were the most relaxing. I actually found the sounds themselves to give me a headache after a while.

[rich decibels]

I don't need precise frequency control so much as precise measurement. I'm convinced there's a simple CMOS way to build a frequency offset counter, I just can't quite bring it together in my head. Something like an up/down counter that increments on one signal and decrements on another. I'll do some simulations and post any progress...