I built a multiple feedback 2kHz band-pass filter with gain Vo/Vi = 4 and an envelope detector for this filter as shown in this great video tutorial EEVblog #490 - Peak Detector Circuit

When i applied a 1v 2kHz square wave signal at the input of the BP-filter i got 4v peak to peak sine wave signal (2v positive and 2v negative).So far so good.

For the peak detector, in the beginning, i chose two OPA379 op-amps.

When i connected the output of the filter to the input of the peak detector and measured the filter's output and the demodulated voltage with the oscilloscope something strange happened.

The filter's output was at 2v (positive peak) as i expected but The demodulated detector's output voltage was 1v instead of 2v (exactly half of what i expected, since the detector holds the peak voltage). Then i changed the two OPA379 op-amps with two OPA340 op-amps and the detector's output voltage was at 2v (as expected in the first place). So i cannot understand why this thing happens. Both OPA379 and OPA340 (or OPA350) are single supply cmos op-amps. Why did the OPA340/50 worked but the OPA379 didn't? Ι assume that i have disregarded something in the datasheet but i don't know what this can be.

I attach two pictures of the faulty one and the correct one (same circuit with different peak detector op-amps).

Thank you

Circuit with opa379 Circuit with opa340

  • \$\begingroup\$ According to your scope traces, in the first instance the BPF input is about 0.5 vpp, while the second trace shows BPF at 1 vpp. Why your input buffer is behaving like that, I have no idea, but it's not the envelope detector that's misbehaving. \$\endgroup\$ Feb 22, 2016 at 1:49
  • \$\begingroup\$ The voltage scale for the first circuit is 2v/div and voltage p-p is 0.5v and in the second circuit the voltage scale is 1v/div and voltage p-p is 1v. I just didn't set the oscilloscope at the same voltage/div scale. It is not wrong. please check it \$\endgroup\$
    – Sparky
    Feb 22, 2016 at 10:06
  • \$\begingroup\$ Some op-amps have input to input ESD protection diodes - maybe this is the problem? When operating as a peak detector there will always be a the requirement to NOT have back to back internal diodes across the inputs. \$\endgroup\$
    – Andy aka
    Feb 22, 2016 at 18:45
  • \$\begingroup\$ I don't understand that. How can the protection diodes change the output? Thank you \$\endgroup\$
    – Sparky
    Feb 23, 2016 at 9:19
  • \$\begingroup\$ Nobody for the rescue? \$\endgroup\$
    – Sparky
    Feb 29, 2016 at 12:09


Your Answer

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

Browse other questions tagged or ask your own question.