I have reproduced a circuit proposed by Nicholas Burnett in 2013:

An improved noninvasive method for measuring heartbeat of intertidal animals please see Fig. 2

which looks as follows: enter image description here

The circuit heavily reacts if I move my hand around it (also pretty far away), which can be seen by the indicator LED. It's a sign for me that it kind of works. However, it seems to be very noisy to me.

Is there a way to make this amplifier circuit more stable either by electronic parts, shielding or grounding it somehow?

EDIT1: The circuit was built as a PCB (source: Chinese manufacturer), please see a screenshot from EAGLE below as well as a photo: enter image description here enter image description here

  • 1
    \$\begingroup\$ This circuit has quite a high gain (over a fairly narrow bandwidth) and power supply may be noisy and (not shown) construction might not be ideal. OTOH, it may be working perfectly when you strap it appropriately to a mollusk. I'm not sure there is enough of a specific question here for a useful answer. \$\endgroup\$ May 4 at 7:11
  • \$\begingroup\$ @Marcobobinger: Does the output change when you wave your hand around the circuit, or when you wave your hand in front of the CNY70? \$\endgroup\$
    – JRE
    May 4 at 7:31
  • \$\begingroup\$ @SpehroPefhany I fear my question might really not be specific enough but I'm a bit puzzled how to troubleshoot this from here on. \$\endgroup\$ May 4 at 14:22
  • \$\begingroup\$ @JRE the circuit reacts on both, around and in front of, probably around is not a good sign as it is unlikely to come from the reflected IR signal? \$\endgroup\$ May 4 at 14:23
  • \$\begingroup\$ What leads you to conclude that there is trouble? Does moving your hand around it influence it simply from the light sensor or from some other factor? \$\endgroup\$ May 4 at 14:24

First of all, every electronic circuit in this universe has, among others, a property called Electric susceptibility.

Second of all, your circuit has quite a high sensitivity giving that it reacts with your hand quite some distance away. It can also be said that your circuit is a highly sensitive antenna. Keep in mind that your signal is amplified by your amplifier, which looks like it amplifies even the smallest signals. If you build your circuit using a breadboard then it is highly susceptible to small signals.

I can recommend three things to do:

  • use strong connections between components
  • build a PCB
  • reduce amplifier sensitivity


So I looked again at your circuit and I can see the POT with a 2MO resistance, having 1 pin left unconnected. You have 2 stages of differential amplifiers. In this case, varistor behavior is unpredictable which can be at maximum resistance thus gain is quite high. So besides strong connections, stable power supply, and so on, I'd also connect pin #3 on the POT to GND. You're actually modifying second-stage gain with that varistor.

  • \$\begingroup\$ I have already tested the circuit on breadboard and the behavior is totally different from the PCB I have built, the PCB seems to work better (the breadboard didn't have much sensitivity at all and showed no clear reaction). The PCB at least reacts to movements of my hand. I'll edit my question with a screenshot of the PCB. \$\endgroup\$ May 4 at 14:28
  • \$\begingroup\$ Please connect the floating POT pin #3 to GND \$\endgroup\$
    – andrew
    May 4 at 18:33
  • 1
    \$\begingroup\$ Thank you, reducing amplifier sensitivity and grounding the floating pin helped. \$\endgroup\$ May 6 at 14:42
  • \$\begingroup\$ @MarcoBobinger excellent. What are your results now? \$\endgroup\$
    – andrew
    May 6 at 20:03

I would try a 10n or so capacitor from pin 3 of the op-amp to ground. The LM358 inputs can rectify RF.

Also make sure at least the power supply you are using has a grounded output.

Kind of a strange circuit, they're using the Zeners as regular diodes (because they had them kicking around, most likely).

When you get closer to getting this working you may need to adjust the value of R2 to bias it properly, there is no compensation for CTR of the opto. The symptom would be insufficient sensitivity. Check the phototransistor output with a voltmeter.

  • \$\begingroup\$ I reproduced the circuit with Zener diodes but also didn't see a reason why you shouldn't use standard silicon ones. I'll try the 10n solution and report back. The higher R2 the higher the sensitivity? \$\endgroup\$ May 4 at 14:56
  • \$\begingroup\$ It's more important to bias it so that the voltage is somewhere in the linear range (not too close to 0V or 5V). \$\endgroup\$ May 4 at 15:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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