Stupid Q: easy two-ocatve up circuit

[markusw]

I was thinking of driving the envelope follower of VCF with the signal that is pitched two octaves up in order to reduce ripple. Is there an easy solution for this?

Thanks,

Markus

[R.G.]

Yes, and no.
No, not the way you're thinking of it. Generating a signal two octaves up that is the same size as the signal you start with is very, very hard.

However, it's common in the synthesizer world to full wave rectify a signal, not filter it to DC, then AC coupling the FWR signal to remove the DC component, and full wave rectifying the signal again. This double FWR signal is then used to make the envelope. It works well, just takes a lot of parts to do.

[Paul Perry (Frostwave)]

RG is right, it is common, and there is a classic old Elektor design somewhere that used a pair of NE570 or maybe NE571 to do this. But, I personally never found any noticeable qadvantage, at least if the envelope was going to controll the signal it was derived from.

[markusw]

Thanks a lot guys!

But, I personally never found any noticeable qadvantage, at least if the envelope was going to controll the signal it was derived from.

Not even with bass guitar (I play a 5-string)??

and there is a classic old Elektor design somewhere

Any ideas where to find it?

Generating a signal two octaves up that is the same size as the signal you start with is very, very hard.

Could it be done by full wave rectify the signal, amplify it through an op amp and full wave rectify it again?

[Paul Perry (Frostwave)]

Generating a signal two octaves up that is the same size as the signal you start with is very, very hard.

Could it be done by full wave rectify the signal, amplify it through an op amp and full wave rectify it again?

...it will be a terrible mess, unless it is a single note.

[R.G.]

Quote:
Generating a signal two octaves up that is the same size as the signal you start with is very, very hard.

Could it be done by full wave rectify the signal, amplify it through an op amp and full wave rectify it again?

That is the double-FWR system I described.

From your description I took it that you wanted to do some actual octave-up operation to generate a signal two octaves up. FWR generates a stew of distortion products, and it's not really much of a signal generation operation. There are few ways to generate an accurate octave up signal, usually DSP or something like the Mu-Doubler. DSP is the best bet, but if you're doing something that complicated, you're 'way beyond the complexity you are talking about. The Mu-doubler will not preserve the original signal loudness very well, and that's where you were trying to go.  There are multiplier circuits, but they need trimming to work well. That's why I said that the operation was hard  - unless you use a double FWR.

Since the NE570/571 is a dual circuit, you can use a single IC to do the double FWR. But they're hard to find.

There's another way. There is a really good envelope detector design using CMOS logic and opamp peak detectors. See the second section in, Harry Bissell's envelope detector. It's a really good way to follow an envelope in a frequency-sensitive way. Go to http://www.reed-electronics.com/ednmag/contents/images/122602di.pdf

[toneman]

hey guys,
which Elector article R U refering 2??
the 570/571 is a compressor/expander.
how do U do an octave or 2 up with a compressor?
EML used a PLL to generate a higher clock freq
to clock a TopOctaveGenerator for doing chords
from single synthesizer notes.
RG, Paul, do either of U have the Elector article??
i've always done freq doubling with E-OR gates.
It comes out a pulse wave, and U can only do it
twice, or double the input freq,
sure would like 2 check it out...
afn
tone

[Paul Perry (Frostwave)]

the 570/571 is a compressor/expander.
how do U do an octave or 2 up with a compressor?
tone

they have a fullwave rectifier as part of the chip,  this is used (unfiltered) for the double rectification.

[puretube]

I don`t think the 570/571 are hard to find...

cc. "Elektor": I don`t remember a double-fullwave rectifier circuit there in the last 34 years; (lots of env-foll., though...)

?

[markusw]

Actually I would have preferred to generate an actual octave-up operation, but I could not find an easy circuit for it. Now its a little more clear why. I have seen the Harry Bissel filter, but I thought it could be easier two simply go two octaves up.
One more Q: how do you

.. then AC coupling the FWR signal to remove the DC component...

:?:  :?:

[R.G.]

One more Q: how do you

You put in a series capacitor large enough to pass the lowest signal you will want to pass. This series capacitor blocks the rectified DC level, which will confuse things in the second FWR.

I don`t think the 570/571 are hard to find...

They may not be there, but they are here in the USA.

[Paul Perry (Frostwave)]

Steve at smallbear electronics has the SE571, identical to the NE571.

[markusw]

..it will be a terrible mess, unless it is a single note.

Why?

You put in a series capacitor large enough to pass the lowest signal you will want to pass. This series capacitor blocks the rectified DC level, which will confuse things in the second FWR.

It also confuses me :?  How can I rectify a signal thats already rectified?

[brett]

Gees, I was a bit lost in the detail.  But "rectify a rectified signal" is easy :- pass it through a cap so that it's 50:50 positive:negative.

[markusw]

..pass it through a cap so that it's 50:50 positive:negative

Thanks. I had a knot in my brain.

[Paul Perry (Frostwave)]

cc. "Elektor": I don`t remember a double-fullwave rectifier circuit there in the last 34 years; (lots of env-foll., though...)
?

I have a very blurry print of a (now lost) scan of what is definitely a German  elektor envelope follower (or hullkurvextraktor, ?spelling) using the two rectifiers in a SE571 in series. I airmailed it today to another listmember, maybe he can pass it on to someone for a redraw.