ECG circuit help [no output at notch filter]

[nuclearrambo]

I have constructed an ECG circuit but I am unable to get the output.

Here is what happens:
I get a noisy output from INA128, I filter it with the band pass filter. I get output from the bandpass filter too. Everything is fine until here.

I do not get any signal at the output of notch filter nor do I get any output from the final inverting amplifier. What could be wrong?

The circuit is attached below:

Also this is the output that I get after processing the signal in MATLAB

 

[godfreyl]

That PDF is a complete mess. Seriously. You've cut U3B right in half, with the top half on page 2 and the bottom half on page7, with other bits of circuit and a couple of blank pages in between.:shock:

Anyway, the important question is: What is the input of the twin-T notch filter connected to?

I see two possibilities:

1. The output of U1D, which is the output of the bandpass filter. That makes sense and should work.
2. The output of U3B, which is set to a constant voltage half way between VCC and VEE. That would give no output.

 

[nuclearrambo]

U3B is used to create a virtual ground.

The input to twin T notch filter is the output of bandpass filter.

I am sorry about the pdf. I have uploaded a new file.

 

[godfreyl]

Maybe something's connected wrong. I'd check if there's any signal at the input to the notch filter.

 

[D.A.(Tony)Stewart]

twin T is overloaded.. Connect it to + input. and R from - to virtual ground.

Also Q if Notch is determined by mismatch of parts, so don't expect too high with this component selection.

<1% desired.

 

[nuclearrambo]

Maybe something's connected wrong. I'd check if there's any signal at the input to the notch filter.

Yes there is signal at input of notch

twin T is overloaded.. Connect it to + input. and R from - to virtual ground.

Also Q if Notch is determined by mismatch of parts, so don't expect too high with this component selection.

<1% desired.

I didnt quite understand what you meant by the first line.

did you mean to say that the amplifier at the output stage should be in non inverting mode?

 

[D.A.(Tony)Stewart]

twin T is overloaded.. Connect it to + input YES NON INVERTING the inverting input impedance is Rin needs to be much higher if used.

The design has many other opportunities for noise reduction. The twin T is easily good for 40~50dB hum reduction.

Your virtual ground impedance is too high (1k) and input is unfiltered.

 

[nuclearrambo]

twin T is overloaded.. Connect it to + input YES NON INVERTING the inverting input impedance is Rin needs to be much higher if used.

The design has many other opportunities for noise reduction. The twin T is easily good for 40~50dB hum reduction.

Your virtual ground impedance is too high (1k) and input is unfiltered.

Should I remove the 1k resistor from V. GND circuit?
For filtering should I add a large value capacitor at the power supply input?

About the non inverting part: I had already tried previously with non inverting, but I had the same problem. so I switched to inverting mode and the problem persists.

 

[D.A.(Tony)Stewart]

Since you already have a split supply, the virtual ground is redundant and worse noise noise immunity., so disconnect U3B and use GND.. All Op Amps ought to be 0Vdc out.

 

[nuclearrambo]

Since you already have a split supply, the virtual ground is redundant and worse noise noise immunity., so disconnect U3B and use GND.. All Op Amps ought to be 0Vdc out.

No I plan to use this circuit on a single 9V supply. That is the reason I included a virtual ground circuit here.

 

[D.A.(Tony)Stewart]

Then correct your wiring and schematic. All OA's should be at 4.5Vdc

report back Vin too, if not.

Yes change 1K to 0R and add 0.1uF is optional on input.
All wires should be shielded or at least twisted pair with the ground. High u CM choke will help suppress hum as well.