I am trying to build a circuit capable of measuring ECG signals following the scheme attached. enter image description here

Since AD623 was out of stock, i ended up buying something like this: https://www.arduino-tech.com/cjmcu-8221-ad8221ar-precision-instrumentation-amplifier-module/.

My concern is now: how should i or better said, where should i connect the GND pin from the module?

As far as i can tell, after analyzing the breakout board, the module comes equipped with the input resistances, gain resistance and the capacitors outlined by me in the red square. I followed the presented scheme accordingly, built the driven ground circuit and connected the +VSS pin to the V+ output, the -VSS pin to the V- output and the GND pin to the AGND. Afterwards, since i don't want diode protection for the time being and since the previous assumption, that the module already has the input resistances, i took the 47K ones and put them on the left of the 1M ones with the view of acting as protection resistors. Everything else was kept intact, i only disconnected the right driven leg circuit, because i also don't need it from now, but i was unable to obtain any output.

I should mention that i tried to isolate only the INA block to see if it works, and this is ,i think, where the problem is located. The datasheet (https://www.analog.com/media/en/technical-documentation/data-sheets/AD8221.pdf) specifies to tie REF to the appropriate local ground, so i assumed it must be linked with the AGND. So this is my configuration at the time being: -VSS pin to V-; GND pin to AGND; +VSS pin to V+; REF pin to AGND. On the IN+/IN- i tried to gather signal both from electrodes located on the skin and from a 5khz generator but i got nothing. The generator was connected with its ground to AGND, and with its output to one 47K resistor and then on both 47k resistors tied together and no result. The only concerning result i had was the fact that at one point i was reading from the output a 50Hz signal, whereas both my circuit and the generator where powered from 9V batteries, and the oscilloscope, a Hantek 6022 one, was connected to the PC, which was not plugged into the power-line socket.

So, again, what i am asking is: How to properly connect the breakout board, since there must be a problem with the AGND driven ground potential and the mismatch with the REF/GND pins from the module. (I will also recheck my soldering job for any discontinuity, just to be sure.)

Thank you for your answers!


1 Answer 1


The ground pin on the module is for the bypass caps (as far as I can tell by looking at the board)

The REF pin is for a voltage offset. For example if you want a 2.5V offset, you could tie the REF pin to 2.5V and the output would be shifted up by 2.5V

In your application the REF isn't needed (except for the low frequency attenuation circuit) to provide a DC offset because the DC offset is blocked by capacitors upstream.

One thing that might be good to check is the voltages on the AD8221's PCB to make sure they are correct.

You should be able to power on the AD8221 with the 9V batteries. Check the amplifier to make sure it is working by a circuit like this:


simulate this circuit – Schematic created using CircuitLab

Move the rail within the input voltage range (output range divided by the gain) with the potentiometer and you should see the output change also. If not then the amp might be broken

  • \$\begingroup\$ Hello and thank you for the reply! \$\endgroup\$
    – Steven
    Jul 24, 2019 at 21:53
  • \$\begingroup\$ I believe the amp is functioning well, no doubt in that, since it amplifies the 50 Hz noise; i tried a different experiment: i powered the INA from a +12V power supply as follows: -VSS to ground (the power supply one), +VSS to +12V, GND from the breakout board to the power supply ground and by your recomendation, let the REF pin floating. No result on the output, except the 50 HZ noise mentioned. How should i tie the GND pin? Any info would be much appreciated. (And even if i would like to try your suggested configuration, i should tie the GND pin to the ground of the scheme?) \$\endgroup\$
    – Steven
    Jul 24, 2019 at 22:00
  • \$\begingroup\$ Yeah, well i figured out a first mistake: AD8221 needs a differential supply in order to function :D \$\endgroup\$
    – Steven
    Jul 24, 2019 at 22:20
  • \$\begingroup\$ I've tried another test that did not work...Could you please advise how to connect the pins from the breakout in order to properly test your proposed function? (I assume it must be something in this manner: +VSS to +V1, - VSS to -V2, -V1, +V2 and IN- to GND, IN+ to the potentiometer, potentiometer also to +V1 and -V2, REF floating. Is this ok?) Thank you! \$\endgroup\$
    – Steven
    Jul 24, 2019 at 22:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.