Need explanation with the working of analog filtering.

I used the analog circuit of two satge filtering as shown in this link https://embedded-lab.com/blog/?p=5508and got the output signals as shown here




with the yellow representing the first stage output and blue represents the second stage output.
As from the output image signal it can be seen that the amplitude of second stage starts increasing slowly when the first stage signal (or the input to the second stage) starts increasing and attains a peak value when the amplitude of input signal(first stage output) is maximum and slowly drops to zero value. This starts again when the input signal (output of first stage) starts increasing from a zero value.
Can someone explain how the second stage works or changes the pulse signal (output of the first stage fed as the input to it) to a nearly square wave.
thanks.

I don't see anything that looks like a "nearly square wave".
And I think the blue is the first stage output and the yellow is the second stage output.
But the second stage is just a low-pass filter which rounds off the high frequency peak from the first stage and stretches it out to give the rounded signal. Nothing unusual there. It's a standard characteristic of a simple RC low-pass filter.

The circuits he's shown use an opamp with a single supply rail and no bias, so they clip badly, cutting off the bottom half of the waveform.

See his previous dozen threads on the same subject.

godfreyl said:

The circuits he's shown use an opamp with a single supply rail and no bias, so they clip badly, ...........................

Click to expand...

Actually, it would appear that they clip quite well. :wink:

LPF is a sort of integrator. In this case it acts like capacitor connected after rectifier. It smooths pulsating voltage. For more explanation we need complete circuit.
I suppose that shape of pulses exiting first stage depends on force you are pressing a sensor (square or like a sine wave).

The yellow is the optical heartbeat from the output of the first unbiased lowpass filter opamp and the blue is the output of the second unbiased lowpass filter opamp with a gain of 101 times. His output signal levels are not as high as in the original project maybe because he is using a different colored LED and a different light sensor. The circuit has coupling capacitors that cut low frequencies and integrator capacitors that cut high frequencies. Here is the optical heartbeat from the article:

@godfreyl :can I know what do you mean by "clip badly, cutting off the bottom half of the waveform".
Is there is a possibility to change the design so that it can work better without cutting of the bottom half of waveform.

@audioguru:yes, I guess you are right that the difference might be because of usage of different LED and a different light sensor.
But doesn't the combination of RC make a highpass cutting the low frequencies and the active lowpass with opamp cuts off the high frequency.
can I get some explanation of this "The circuit has coupling capacitors that cut low frequencies and integrator capacitors that cut high frequencies".

If you bias the inputs of the opamps at half the supply voltage then their outputs can swing up and down (without cutting off the bottom half of the waveform) the same as a heartbeat signal. But then maybe you cannot feed the signal to a microcontroller because the signal will not swing all the way down to ground.

A coupling capacitor feeding a resistor to ground is a highpass filter (it cuts low frequencies). A capacitor (integrator capacitor) parallel to the negative feedback resistor of an opamp makes a lowpass filter (it cuts high frequencies).