[SOLVED] ECG signal circuit issue and noise interference

Hello everyone.
I have designed one proof of concept for a TWO electrode method for ECG. image below is the schematic.


Power supply is taken from li-ion battery pack of 11.1 V made by 3 cells in series. For 5V I have used one LM1117 or LM7805. AGND reference is taken fro voltage divider circuit passed through a buffer to get +2.5V and -2.5V from 5V itself.

I facing two major issue

1). There is 50 Hz interference which is overlapping with ECG signal.

2). There is a swing in ECG signal i.e. the ECG signal is going beyond the supply voltage range (0 - 5v).

Here is some screenshots of the signal taken in DSO.

Below, signal taken from stainless steel plate and thumb.

Below, FFT showed 50Hz interference without any input to the instrumentation AMP.


Below, Signal going beyond the supply voltage and can not be measured.


I not able to understand the error of 50Hz interference even after using battery power supply. And why the signal going beyond the reference and power supply.

Please help me figure out cause of this issue and suggest some diagnosis method to solve the circuit issue. And if the question lags some information please do let me know. Also suggest me if there is any digital solution that can be implemented to get proper result.

Thank you.

Hi,

There are many ECG signal conditioning circuits in the internet, even in the forum I remember some discussions.

50Hz is a known problem. A notch filter could help.
Suitable filters are essential to get good quality signal.
You have to decide upper and lower frequency of interest.

Regarding "signal beyond supply rails"...
I'd reduce the two 1MOhms to 100kOhms.

And instead of connecting them to GND I'd bias them to VCC/2.
Just use two additional resistors and a capacitor.

Klaus

I think shielded audio cables should be used and most ECG circuits add the two inputs (making a common mode signal), invert it and feed it to the patient's right leg, to cancel the AC and DC interference:

You are pumping any noise on your 5V supply straight into you amplifier through that top capacitor. You also have not established your DC level relative to the human body. Also if this is going to be used in a real medical instrument you need to meet 60601-1 patient isolation requirements for safety. You also have things like a capacitor from ground to ground which makes no sense. Label the components (R1, C2, etc) and this can be discussed.

KlausST said:

Hi,

There are many ECG signal conditioning circuits in the internet, even in the forum I remember some discussions.

50Hz is a known problem. A notch filter could help.
Suitable filters are essential to get good quality signal.
You have to decide upper and lower frequency of interest.

Regarding "signal beyond supply rails"...
I'd reduce the two 1MOhms to 100kOhms.

And instead of connecting them to GND I'd bias them to VCC/2.
Just use two additional resistors and a capacitor.

Klaus

Click to expand...

Thank you so much. The signal is coming perfect now. Noise is also reduced by introducing a filter after the inst. amp. I have also shielded the cable to reduce 50Hz nearby interference.

Sorry for the late response.

And thank you to every other member for your share of knowledge. I really appreciate.