# A few questions regarding the functionality of this negative peak audio detector

I have some queries regarding this peak detector from an audio compressor circuit:

My understanding of the process is that when the positive input swings low, D5 is negatively biased and stops conducting whereas D4 is positively biased so the input of the op-amp sinks current and charges C1 down to a negative voltage. During this period there is no negative feedback so the output of the op-amp goes into negative saturation.

Once the voltage across the capacitor becomes equal to the input voltage, the inputs of the op-amp become equal and the op-amp stabilises. The image below shows a simulation of the circuit in which the input voltage drops from 0, to -1V. Why does the output of the op-amp (purple trace) stabilise at -0.6V?. I note that this is equal to one diode drop voltage, but can't see why this happens.

Also, it's not visible on this image but over time the op-amp output slowly drops down to become equal to the input voltage, but the circuit is static, so why would there be current flowing to cause the voltage to drop?

Finally, given that this is a negative peak detector, aren't these capacitors in the wrong direction? If they're being charged to negative voltages, shouldn't the anode be connected to ground?

• Consider the charge (and hence voltage) on C4.
– user16324
Commented Feb 22, 2021 at 13:15
• I can see that C4 can discharge via the resistor in the feedback loop when the op-amp value is high enough to bias D5. But in the simulation the voltage never increases? Commented Feb 22, 2021 at 13:25

Why does the output of the op-amp (purple trace) [initially] stabilise at -0.6V?

The feedback loop (R at the top) compensates for the diode drop.

Why does the output finally go to zero?

This will depend on the diode model used. A good model will conduct some current at very low voltages and eventually have a drop of zero.

Are the caps in the wrong direction?

Non-polarized caps would normally be used here (better tolerances). I realize that you didn't draw this, but when drawing your own schematics, please don't use that symbol for polarized caps, it is not widely used anymore and not everyone will immediately recognize it. Add a plus sign for polarized caps.

What is the purpose of D5?

D5 keeps the opamp out of saturation (when the input is + for a negative peak detector). It takes time to recover from saturation and part of the next valid input signal can be missed without it.