Electronic stethoscope

[audioguru]

The second opamp is the Sallen and Key lowpass filter. It has a gain of 1.6 (33k/56k plus 1= 1.59) so that it has a Butterworth flat passband then sharp corner response. With a little more gain (2.2) it will have a peak at one frequency. If its gain is more than 3 then it oscillates.

The first opamp can have any amount of gain you want. Now its gain is about 10. If the 47k resistor is increased to 470k then its gain is 100.

 

[Willbe]

If you only need to count, you could saturate the amp and lose the heartbeat sound quality.
I expect your range will be 40 to 220 beats/min (for adults).

FWIW I remember a circuit for doing this that measured the period and then calculated the inverse. I forget what the advantage was to this method; it might have been to improve the S/N ratio.

 

[microtexan]

Maybe I’m looking at it wrong, but most are looking at detecting an audio signal (the heart beat) and changing that to a pulse to count. Why not just detect the heart beat electrical signal (like an EKG does) and count that. Amplifying the small voltage would be about as easy and would not involve the noise an audio signal does.

 

[audioguru]

An EkG is different from a stethoscope.
The EkG picks up the brain's electrical signal that drives the heat.
The stethoscope picks up the sound of the heart and all other noises around.

 

[microtexan]

stethoscope

An EkG is different from a stethoscope.
The EkG picks up the brain's electrical signal that drives the heat.
The stethoscope picks up the sound of the heart and all other noises around.

But the OP said he was just interested in counting the pulses with a uC. Why not just count the electrical pulse instead of converting the sound to a "pulse" and then counting that?

 

[NeoMach]

An idea

Use an inverting op-amp on different location where pulses are not as strong, and feed the two signals into an adder. The inverted noise should cancel the noise on the pulse sensor, and you will be left with the pulse.