Peak rectifier problem

[xmat]

Hello.

I am looking for a peak rectifier circuit to rectify low frequency audio signals..however it is a little trickier than just a PR circuit.

I want the circuit to work only when the signal is decreasing.

Meaning that a normal PR would (let's say) include a capacitor with a charge and discharge behavior..
I -on the other hand- would like a pr that does not work (is off) during discharge times..
so I can only peak rectify the rising parts of a singal.

Is it possible?

Hope somebody can help,

thanx in advance,

xmat.

 

[Roff]

Can you draw a waveform of what you want? I don't get it.

 

[xmat]

Can you draw a waveform of what you want? I don't get it.

Ok.

Here is a schematic with the signal I am trying to peak rectify..(Both blue and green consist the signal,
while blue is the part I want to rectify and green the part of the signal I don't want to peak-rectify..)

All I want is to rectify the info of the signal when it is rising, while having no rectification of the signal when the waveform is going downwards..See schematic.

Think of it as a "selective peak rectifier"..Rectifies only the "rising" parts of the signal.

Also, if this is achievable, can I do the opposite thing, i.e rectify only the falling parts of the signal keeping the rising parts unaltered.

Thanx for your help,

xmat.

 

[Russlk]

Am I correct that the red line represents the output waveform (my drawing in Paint is shaky)? I don't know how to do that.
Well, I guess a peak detector and sample and hold would do it.

 

[Nigel Goodwin]

As with many of these questions, perhaps you should tell us what you are trying to do?, it might be (and often is!) that there may be a simpler approach?.

 

[zevon8]

Perhaps a digital VU meter circuit, and you could adjust the "decay time" to suit the frequencies involved.

 

[xmat]

As with many of these questions, perhaps you should tell us what you are trying to do?, it might be (and often is!) that there may be a simpler approach?.

Well, it is not always that simple..I don't want to bore the hell out of people with a 2000 words' post..

Anyway, for your information, I am trying to "selectively" peak rectify (rising or falling) parts of this waveform to extract from it information about the rising time-of-change of the signal, or the falling time-of change of it.

Main reason for that: I want to extract from the signal the "basic" peak, (meaning I have to filter all others), leaving its rising and falling characteristics INTACT...Got it?
(The picture I send u is ONE signal..The main problem with a common peak rectifier is that it cannot output the main/larger peak without altering its rising/falling characteristics..That is rate of change, I guess)

Maybe a slew rate block would do the trick..What'd u say?

Thanx for your willingness to help,

xmat.

 

[Roff]

<snip>I want to extract from the signal the "basic" peak, (meaning I have to filter all others), leaving its rising and falling characteristics INTACT...Got it?<snip>

No. :?
You need to show a waveform of input and output. I'm assuming you know what you want, but you just don't know how to do it.

 

[JimB]

I have a bad case of deja vu here.
I remember some weeks ago there was a thread on a very similar topic, a peak detector that had to preserve some other information to allow the signal to be re-contituted afterwards (or something like that).
To tell the truth I did not understand the problem then, and I dont understand it now.

JimB

 

[Russlk]

Perhaps a diferentiator circuit, where the output is the rate of change of the input is what you need? The rising edge will give a positive output and the falling edge will give a negative output. The amplitude of the output will be proportional to the rate of change.

 

[xmat]

Sorry, guys..Multiple posts due to a short loss of memory..(need vacations)..See the following thread to understand what I am trying to do.

https://www.electro-tech-online.com/threads/voltage-regulator.17464/

xmat.

PS:Ron H, a usefull schematic with input and output graphics for you to disolve your curiosity.

 

[Nigel Goodwin]

Sorry, guys..Multiple posts due to a short loss of memory..(need vacations)..See the following thread to understand what I am trying to do.

https://www.electro-tech-online.com/threads/voltage-regulator.17464/
.

That still doesn't explain what you want to do!.

 

[xmat]

Sorry, guys..Multiple posts due to a short loss of memory..(need vacations)..See the following thread to understand what I am trying to do.

https://www.electro-tech-online.com/threads/voltage-regulator.17464/
.

That still doesn't explain what you want to do!.

Detect occurence of (impulsive) peaks within a noisy background, by using rate of change of signal as the parameter for detection, of let's say musical beats (beat detection)..
The signal you see is ONE signal, and I have to find I way to make a circuit understand it is one and not more...by detecting the fastest peak and passing it through unaltered while filtering extra peaks..Now?

 

[Nigel Goodwin]

of let's say musical beats (beat detection)..

Simple! - use a low pass filter, it's a very common technique!.

But I suspect that isn't what you've been asking?, because it bears no resemblance to anything else you've asked!.

 

[xmat]

of let's say musical beats (beat detection)..

Simple! - use a low pass filter, it's a very common technique!.

But I suspect that isn't what you've been asking?, because it bears no resemblance to anything else you've asked!.

Really???That simple???

I should inform you that such technique has a limited dynamic range, i.e it totally filters out weak signals; or if u "tune" the filter to "hear" those weak signals, u get wrong beat detection during high level ones (level variation also changes frequency content, which enhances the potential for false detection..)...

What makes u think I NOT am saying what I am trying to do?

xmat

 

[JimB]

Having looked at:
https://www.electro-tech-online.com/threads/voltage-regulator.17464/
I think I understand what you are trying to do.

You have a randomly varying signal and you are trying to detect and store transients/pulses from that signal which have an amplitude greater than the average amplitude of the signal.
Also, you only want to detect and store the first pulse of a series of pulses.

Am I correct with the above statement?


In your graphics in https://www.electro-tech-online.com/threads/voltage-regulator.17464/, there is a bit of a confuser.
The pulse on the right hand side, is that the first pulse and the pulses "A" come after it?
If so the first pulse should surely be at the left hand side, time advances to the right on a conventional oscilloscope or trend recorder.

JimB

 

[xmat]

Having looked at:
https://www.electro-tech-online.com/threads/voltage-regulator.17464/
I think I understand what you are trying to do.

You have a randomly varying signal and you are trying to detect and store transients/pulses from that signal which have an amplitude greater than the average amplitude of the signal.
Also, you only want to detect and store the first pulse of a series of pulses.

Am I correct with the above statement?


In your graphics in https://www.electro-tech-online.com/threads/voltage-regulator.17464/, there is a bit of a confuser.
The pulse on the right hand side, is that the first pulse and the pulses "A" come after it?
If so the first pulse should surely be at the left hand side, time advances to the right on a conventional oscilloscope or trend recorder.

JimB

"....The pulse on the right hand side, is that the first pulse and the pulses "A" come after it?...."


Well, no...That's the mind numbing about it and that's the reason I asked for your help (which is greatly appreciated)!

As I already mentioned, this is one sound.

The strange thing about this kind of sound is that its main peak/transient may appear LAST and not first (at least 5% of all times).

Ahhh, I forgot to mention that the signal does not end at the 10th millisecond, it continues (just did not draw it) with a very slow decay envelope...the signal last around 50-100ms.

"....You have a randomly varying signal and you are trying to detect and store transients/pulses from that signal which have an amplitude greater than the average amplitude of the signal..."

EXACTLY.

One comment to add is that I want to detect the greater transient- the one with the higher rate of change.All others (slower) I want them filtered/smoothened-out.

And I am trying to do that by filtering slow-varying parts of the signal (filtering means making them slower) while letting rapid-varying parts of the signal pass unaltered.

Also, it would help if there was a way to set indepented slew rate control on rising versus falling parts of the signal.

Anyone can share a though with me on a possible schematic?

Thanx in advance,

xmat.

 

[Russlk]

A signal processor could do what you want but programming it is beyond my capabilities. If you can describe the process, it can be done, but you have not convinced me that you are able to describe the process.

 

[Roff]

One comment to add is that I want to detect the greater transient- the one with the higher rate of change.All others (slower) I want them filtered/smoothened-out.

How do you define the time interval for searching for the greatest transient? If you don't define an interval, you have to wait forever. Transients are like earthquakes - there will probably always be a bigger one if you wait long enough.

 

[Nigel Goodwin]

of let's say musical beats (beat detection)..

Simple! - use a low pass filter, it's a very common technique!.

But I suspect that isn't what you've been asking?, because it bears no resemblance to anything else you've asked!.

Really???That simple???

I should inform you that such technique has a limited dynamic range, i.e it totally filters out weak signals; or if u "tune" the filter to "hear" those weak signals, u get wrong beat detection during high level ones (level variation also changes frequency content, which enhances the potential for false detection..)...

It works perfectly well for the 100,000's of units which employ that method?.

What makes u think I NOT am saying what I am trying to do?

Because you have never explained it, and nothing you've previously mentioned makes any suggestion of wanting to do musical beat detection!.