Subject:
|
Re: How can I build sensor to measure eeg
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Mon, 6 Aug 2001 17:26:34 GMT
|
Original-From:
|
John Barnes <BARNES@SENSORS.COMantispam>
|
Viewed:
|
596 times
|
| |
|
|
topas wrote:
> I just have a quick question. Hope you guys could give me some pointers.
>
> I'd like to build a sensor that measures the voltage of the brain wave
> like those EEG machines seen in the hospital. And would like the RCX to
> log the data.
>
> I have scanned thru some sites and they explained that these brain waves
> are about 0 - 200 micro volt in magnitude and have about 30 Hz in BW. Here
> is what I think I should do next. But not sure if it's making sense. (I
> have some experience building electronic kits.)
>
> 1. Build an op-amp circuit to amplify the signal to be (-5) to (+5) volt.
> Is this correct? The gain must be enormous?!?
> 2. With (-5) to (+5) volt range, the RCX should be able to read and map
> into the 0 -1023 value well.
>
> Should I use the power from RCX sensor port to feed the op-amp? With that
> enormous gain, am I damaging RCX in anyway?
>
> I read from the Extreme Mindstorm book and found that the RCX sensor port
> would do the switching between feeding the power to the sensor and reading
> the value from it. Could you guys explain how this would affect the sensor
> implementation?
>
> How can I find out how RXC do the sampling? I guess in this case the 30 BW
> is so low and RCX would have no problem with it right?
>
> Last question, will this approach work? Anything I am missing?
First of all, you'd simplify things if you made your amplifier center
around 2.5 volts rather than 0.
Then your amplified signals would become 2.5 +/- 2.5 which the RCX will
like. And if you
run the amplifier from a stabilized 5 volt supply, its output will never
exceed the 0 - 5 volt
range and thus your RCX will be safe no matter how much gain you set up.
The real problem you'll find is that the eeg voltages are very small. Try
emg first, voltages from
your arm muscles are in the millivolt area. The hospitals and clinics that
use eeg machines
normally have screening in the floor and walls to cut down on external
interference.
One area that is very important to get right is the connection to the skin.
If you are not careful
the voltage generated by acidity in the skin will dominate the readings. If
you keep perfectly
still these voltages will just be DC and will disappear, but if you are
moving, you may have
trouble if you don't use buffered gel and pickup electrodes with noble
metal plating.
In your amplifier design, which will be multistage A/C coupled, make sure
you pay good
attention to overload recovery time, you don't want to be waiting minutes
while all your
capacitors balance.
... and be careful of yourself. It is very easy to give yourself quite a
surprise with a well
connected electrode! The RCX is battery powered and your sensor can run
from the
RCX batteries via the sensor port. Make sure you are not using a Rev. 1.0
RCX with a wall
transformer.
Let us know when you've decoded your brain waves :) I tried it once myself,
and couldn't
make a lot of sense out of it. About the only thing I found I could measure
reliably was the
"e" wave.
JB
|
|
Message is in Reply To:
 | | How can I build sensor to measure eeg
|
| Hi, I just have a quick question. Hope you guys could give me some pointers. I'd like to build a sensor that measures the voltage of the brain wave like those EEG machines seen in the hospital. And would like the RCX to log the data. I have scanned (...) (23 years ago, 6-Aug-01, to lugnet.robotics)
|
5 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
