skin conductance

[Tykulcsar666]

Hello,

I posted a question a while back regarding this simple circuit to measure skin conductance. I've been getting data for the project and it seems to be vaguely measuring conductance, but the data is noisy and the peaks aren't as distinct as they should be. I have an example of some data below.

Other than using better electrodes for sensing, looking to see how I can optimize hardware to eliminate noise. There is a 0.5 Hz low pass filter, but not sure of other ways to make the circuit more robust.

Any help would be very much appreciated!

 

[alec_t]

Add decoupling capacitors to the op-amp supply and to the R1/R2 junction.

 

[Tykulcsar666]

Thank you for the response. How do you choose values of the capacitors?

 

[Tykulcsar666]

And only one capacitor is necessary for the supply? Or one for each op amp

 

[alec_t]

Capacitor value isn't critical. I'd try (a) one 100uF (or something of that order) electrolytic, plus one 0.1uF ceramic across the power supply rails, and (b) another ~100uF electrolytic across R2.
Hopefully your circuit is built on a pcb and not on a breadboard (which is notorious for picking up noise).

 

[Tykulcsar666]

Great thank you. The 100uF and .1uF across the power supply you’re saying in series?

 

[Tykulcsar666]

Also circuit is actually on a breadboard right now

 

[gophert]

Great thank you. The 100uF and .1uF across the power supply you’re saying in series?

Put the power supply decoupling capacitors in parallel.

 

[gophert]

Could you post a screen capture of the schematic from post #1? I can't open that file type.