Opamp Envelope filter circuit?

[carboncomp]

Hi,

Does anyone have a link to a resources to help me understand and build an opamp envelope filter?

Want to play around with an envelope fuzz, but having trouble finding out information on the envelope filter part. 

[antonis]

https://synthnerd.wordpress.com/2016/04/06/envelope-circuits-a-simple-ar-design-using-op-amps/

And a more comprehensive (but discrete) circuit..
https://synthnerd.wordpress.com/2016/06/20/envelope-circuits-a-simple-discrete-ar-design/

Also,


(from Denton J. Dailey - Electronics for Guitarists)

[ElectricDruid]

Hi,

Does anyone have a link to a resources to help me understand and build an opamp envelope filter?

Want to play around with an envelope fuzz, but having trouble finding out information on the envelope filter part.

Envelope filter or envelope fuzz? In my head, those are different things. They both include an envelope follower circuit (like Antonis post links to) but they use it different ways. In a fuzz, you'd use the envelope to re-impose the dynamics that the fuzz removed - e.g. it controls the signal volume via an voltage-controlled amplifier of some type. In an envelope filter, the envelope controls the filter cutoff, sweeping the filter up and down - the classic "Autowah" effect beloved of funk rhythm players (wakkawakkawakka!).

One example of envelope fuzz is the Paia Rocktave (which also includes sub octaves):

http://www.generalguitargadgets.com/effects-projects/octave/rocktave/
https://hammer.ampage.org/files/rocktave.pdf

[antonis]

(wakkawakkawakka!).

Is there any transfer function of that..??

[carboncomp]

Envelope filter or envelope fuzz?

OK, so I really like the Vorg F-502 Warp Sound / Pearl F-602 Warp Sound going into another fuzz for a huge sound.

But not a fan of the always-on Buffer on the Vorg F-502 Warp Sound / Pearl F-602 Warp Sound, so I wanted to learn how to get the same effect but omit the buffer section.

Seen a true bypass version, but wondering if the buffer could be removed entirely and end up with a circuit that has 3 opamps rather than four?

[antonis]

I wanted to learn how to get the same effect but omit the buffer section.

Just make IC1B and IC1A non-inverting..

[carboncomp]

Just make IC1B and IC1A non-inverting..

Sorry, I'm a little slow when it comes to opamps, have a VERY basic understanding of them.

Would Making them non-inverting just change the phase, or am I missing some fundermentals on buffers, as I have seen both inverting and non-inverting buffers and a discussion on this forum I wish I could find that went over where and where not to use non-inverting.

Will non-inverting play nicer with Fuzz after them?

[idy]

I think what the oracle is saying is that the first buffer is there because inverting opamp stages have low input impedance and so will play poorly with any output stage that is not perfect; think wah, fuzz, and anything with a passive volume control right at the end. The buffer at the end of the circuit similarly avoids poor interaction with anything after.

Why do you only want to use three opamps?

Perhaps you would cause less trouble by getting rid of the last buffer instead.

[carboncomp]

Why do you only want to use three opamps?

Perhaps you would cause less trouble by getting rid of the last buffer instead.

That was mainly down to me being an idiot and thinking it was another gain stage, rather than a buffer at both ends, even though it clearly isn't!

Yep, think I'm going to have to breadboard it out and play about with input and output buffer configurations, and what Fuzz I'm putting after it!

If I'm going true bypass, do I just omit these parts?

[idy]

This "Warp Sound" of which you speak, the schematic, that is not envelopey at all, its a %^&*ed wah into diode clipping.... there are envelope fuzzes that give the sound back its envelope after fuzzing it, but that is not one.

And I don't get how its filter works. Pot labeled "envelope." It is a resistor to mid voltage, which is then ac coupled to the next opamp stage, so it should affect nothing? Help.

[carboncomp]

This "Warp Sound" of which you speak, the schematic, that is not envelopey at all, its a %^&*ed wah into diode clipping....

They lied to me!!!!!

[m4268588]

http://sim.okawa-denshi.jp/en/OPtazyuBakeisan.htm

Is this way of drawing more familiar?

[Rob Strand]

Is this way of drawing more familiar?

Delyiannis-Friend Bandpass:



Splitting R1 into a divider is done to set the gain.  R1 of basic circuit = R1a // R1b in reduced gain circuit.

[idy]

Is this way of drawing more familiar?

Not familiar, but more comprehensible. Lots of ways to draw a thing...

[carboncomp]

Delyiannis-Friend Bandpass:

Oh, so it's using a band pass filter rather than an envelope filter?

[antonis]

It's using an Infinite-Gain Multiple-Feedback Band-Pass filter of variable center frequency..

Compare the circuitry around IC1B with the one below:

[ElectricDruid]

I did a quick sim of the MFB bandpass filter from the schematic above. A couple of interesting things come out of it.
Here's the basic response:


As you can see, most of the action is bunched up at one end of the pot. This is one design that *really* benefits from finding a C-taper Rev-Log pot. Here's what happens with the 10KC:


The other thing to notice is the frequency range. It's fairly low. It would certainly be worth experimenting with tweaking those 22n caps. 15n, 10n, 6n8 would all be worth a shot. You'd have to hear it to know - and it'll be a question of taste too. An On-off-on DPDT could give you three ranges, which would increase the flexibility a lot for the addition of one control.

[carboncomp]

As you can see, most of the action is bunched up at one end of the pot. This is one design that *really* benefits from finding a C-taper Rev-Log pot. Here's what happens with the 10KC:

Thank you for taking the time to do that! (what simulator do you use?)

I can confirm most of the effect is on the last quarter or the stock 10K pot.

[antonis]

The other thing to notice is the frequency range. It's fairly low. It would certainly be worth experimenting with tweaking those 22n caps. 15n, 10n, 6n8 would all be worth a shot. You'd have to hear it to know - and it'll be a question of taste too.

Well said, Tom..

Although, by simply lowering R7 "stopper" resistor value (say 100R instead of 270R) should extend frequency range upwards significantly..
(maintaning unaffected Q also..)

[ElectricDruid]

Thank you for taking the time to do that! (what simulator do you use?)

LTSpice. Because it's free, primarily! It's pretty quirky interface-wise, but it works well enough.

I can confirm most of the effect is on the last quarter or the stock 10K pot.

Yeah, it certainly looked like it.

Although, by simply lowering R7 "stopper" resistor value (say 100R instead of 270R) should extend frequency range upwards significantly..

Good point. The sim says 1.3KHz instead of 780Hz. Perhaps that's enough.