Sun, 08.16.26 - Debugging Phase 1

I needed a clean space and didn’t want to clean so I headed to one of my favorite coffee shops for the afternoon to make a debugging plan.

I think my main issue is the backfeed which I’m guessing is biasing my output wrong. I have no idea how or why since the ref pin is measuring at 2.5V just like it should be. While I’m in a clean area with light Jazz to keep me calm, I’m going to make a plan.

Figure out the backfeed issue

I was able to isolate the issue down to the circuit that results in the backfeed. My evidence for the backfeed is that when this circuit is connected up, the power LED on my Arduino turns on even if it’s not connected via USB.

Stuff I’ve already tried

• Switching out the IC
• Making sure power and ground aren’t shorted on the arduino
• Wiring pins 2 & 3 on the IC directly to ground in case the electrodes were picking up some kind of noise
• Tried using pin A1 instead

The Circuit that has doomed me

With JUST this hooked up, I get power into the arduino.

So here are my next steps:

1. Pre connection probe the power supply.
2. With power supply connected to the IC, see what is coming from Dout
3. If we’re getting power from Dout, figure out why
4. If we’re not, swap pins again, something in the pin is bad
5. Make sure we have the right resistor on the input (might need a pull down?)
6. Swap for MSP432 or Uno
7. On a last ditch effort, wire up with inputs and use the oscope to get readouts

Understand the internals of the AD620

Data Sheet

I think I ordered this chip off of ebay years ago when I first started this project and it was obsolete then :O. Now it’s extra obsolete but it’s what I have.

A quick skim through the datasheet doesn’t raise any alarms. I’m within all the operating parameters. I also double checked I did the gain right with 5.1k resistor and that gives me about 10x gain. Not unreasonable for this but definitely won’t work well when I actually connect the electrodes to an input source. It still doesn’t explain the output.

General musings

We don’t really have “waves” in our brain. There is not really a propogation through a medium in a rhythmic manner like the waves we think about crashing on the beach or sound waves. What we’re actually measuring with EEG is the change in the electrical potential on our scalp.

This is yet another instance where I will fight the “war on waves.” The word “wave” is so closely tied to periodic motion and that is IMPRECISE!! There is not a periodic motion that we are measuring that somehow shoots out our scalp and is captured by our sensors.

Instead we measure pulsations of an overall electrical field.

The field has multiple rhythms of pulsation all contained in one signal, like a drum circle. Maybe one person in the drum circle has a nice fast beat while another has a slow beat. Then there’s the little kid who chose the loudest drum to pound her little heart out on.

A trained musician can separate out the beats of different players with their ear. We do the same thing with the pulses in the electrical field detected by the electrodes.