# valley peak detector

Hi all, Any one know how to build a simple valley peak detector using a zener or schottky diode? I can build a peak detector with either fine. But can't seem to get the valleys? For peak detector, I've used something like show here, http://www.allaboutcircuits.com/vol_3/chpt_3/5.html I also place a resistor parallel to the cap for discharging. Again I can't seem to get a valley detector.

thanks Frank

What do you mean by "valleys" Do you mean the maximum negative peak or the lowest voltage that the "rectified" signal fall to.

frank_ard: Hi all, Any one know how to build a simple valley peak detector using a zener or schottky diode? I can build a peak detector with either fine. But can't seem to get the valleys?

Why not invert the signal and use your peak detector...?

Something to try (untested)…

jackrae: What do you mean by "valleys" Do you mean the maximum negative peak or the lowest voltage that the "rectified" signal fall to.

Hi Jack, Yes, I am referring to the negative peaks of a sinusoidal input.

@dlloyd: I'll try that circuit. But, I am not sure that will work because the capacitor needs to discharge since the sinusoid amplitude may change and it needs to follow the negative peaks. Maybe having a resistor connected to ground to the right of the diode may work?

@fungus: That option is good but I cannot invert the signal for what I am doing.

I am not sure that will work because the capacitor needs to discharge since the sinusoid amplitude may change and it needs to follow the negative peaks.

Just an assumption I made because the peak detector sample circuit provided in your link didn't have one. No problem, all that's needed is an additional resistor in parallel with the capacitor. Use values that provide whatever RC time constant you need.

dlloyd:

I am not sure that will work because the capacitor needs to discharge since the sinusoid amplitude may change and it needs to follow the negative peaks.

Just an assumption I made because the peak detector sample circuit provided in your link didn't have one. No problem, all that's needed is an additional resistor in parallel with the capacitor. Use values that provide whatever RC time constant you need.

Hi Dloyd, thanks for the reply. Other things can't do with your suggestion is use a regular diode. I'll use a zener/schotty.

Hi All, I am not limited to the first circuit I posted. It can be a circuit with an op amp, or comparator.

Maybe having a resistor connected to ground to the right of the diode may work?

If you have a dual supply, and you want the signal to start at 0V then quickly go negative to the minimum voltage, then I think this would work. (the resistor is now from ground to anode of diode)

Shouldn't your peak detector work just the same if its negative vs positive, as long as you swap VCC+ with VCC-?

If you aren't using a VCC-, then you need to rectify the signal and use the normal peak detector.

frank_ard: Hi All, I am not limited to the first circuit I posted. It can be a circuit with an op amp, or comparator.

An Op-Amp will invert a signal easily.

A zener diode won't do as it'll fail to conduct until the zener point is reached. You can use any old diode providing you rig it up as a "precision rectifier". see the following circuit http://en.wikipedia.org/wiki/Precision_rectifier which also shows it used as a peak detector, further down the page.

I'd like to have something similar to this, https://www.circuitlab.com/circuit/y453h7/precision-active-peak-detector/ Except with an op-amp 0V-9V, and not inverting the signal.

@Jackrae: Yes, I understand that the zener won't conduct till it reaches it's reverse breakdown voltage. Then I'll have to crank up the voltage or dc offset so that the zener is conducting in its reverse voltage state.

frank_ard:

jackrae: What do you mean by "valleys" Do you mean the maximum negative peak or the lowest voltage that the "rectified" signal fall to.

Hi Jack, Yes, I am referring to the negative peaks of a sinusoidal input.

@dlloyd: I'll try that circuit. But, I am not sure that will work because the capacitor needs to discharge since the sinusoid amplitude may change and it needs to follow the negative peaks. Maybe having a resistor connected to ground to the right of the diode may work?

@fungus: That option is good but I cannot invert the signal for what I am doing.

It didn't work with or without a parallel resistor.

"This circuit can be used to generate negative output voltage if the top plate of capacitor is grounded instead of bottom plate (or if the diode is flipped)." From page 6 here.