Reverse Clipping (peak-detecting) circuit

Thread Starter

stepank

Joined Apr 19, 2018
8
Hello AAC -

I am trying to design a circuit that takes a waveform and only leaves the peak, with some variability of how much of the peak is passed. Not exactly a clipping circuit, not exactly a peak detector.

This is the closest I could get: http://tinyurl.com/y7rbzxt7

But it's not ideal for a variety of reasons. The incoming waveform is 10v, the potentiometer has a range of 0-12v - so I'm not sure why I can't get to a resolution that totally nulls the incoming waveform. I'm also using a buffer for the variable voltage threshold - because I can't seem to get it to work any other way - but I feel like there should be a way around that, right?

Can anyone help set me on the right path?
 

danadak

Joined Mar 10, 2018
4,057
Some confusion, do you want to "gate" a portion, in time, of the waveform
at its peak ? Like +/- 1 uS around the peak value ?

Is waveform periodic ? If so freq range needed to evaluate.

Regards, Dana.
 

crutschow

Joined Mar 14, 2008
34,285
Here's the circuit modified to do what (I think) you want.
The output is shown for several different pot clip-point settings
I also had to add a buffer to the pot output to avoid degrading the signal accuracy.

upload_2018-4-24_19-23-33.png
 
Last edited:

Thread Starter

stepank

Joined Apr 19, 2018
8
Thanks for the responses, let me try to communicate it a little clearer, though - because I think I didn't make it apparent.

The frequency range would be 20-20k, roughly.. and it would operate as a gate, insomuch as everything below the threshold would be 0v...

Crutschow's circuit works really well, but I'd like it to operate with bipolar signals, ultimately - is there a way to get to a similar solution with +/-12v? I feel like I must be missing something simple..
 
Top