4016 switches as amplifiers report

[gez]

Cliff, this thread is over 5 years old.  There was no purpose to this circuit other than to see if it could be done.

I'm surprised no boutique guys have come out with anything using this technique.

If you look at a schematic for the 4016 (check the datasheet), you'll see that a switch comprises three inverters:

http://www.standardics.nxp.com/products/hef/datasheet/hef4016b.pdf

Even though the 'drain' resistor limits channel current to the main switch, you don't have access to these inverters and they all leech current.  In short: current consumption, although not astronomically high, isn't exactly ultra low at 9V, either.  Not really battery friendly and, in all honesty, you might as well use a discrete MOSFET or similar.   If you're using this chip already for other things, it's a useful trick to know about, though.  

The high transconductance makes it suitable for simple T-filter wahs, by the way.  Again, not much point when weighing up the pros and cons...

[gez]

Memory is hazy, the above datasheet is for a buffered version.  Can't remember if that's the norm these days.  Either way, there's at least one inverter leeching current, so not as battery friendly as you think.

[earthtonesaudio]

Cool!  Comparing Gez's schematic with the internal construction, if you ignore the final N-channel transistor, you basically have 3 inverters followed by a P-channel source follower.  So it should act like a normal amplifier.  Adding the N-channel back in, you have a fourth inverting stage which is always "fighting" the P-channel.  Crazy that it works. 

[gez]

Cool!  Comparing Gez's schematic with the internal construction, if you ignore the final N-channel transistor, you basically have 3 inverters followed by a P-channel source follower.  So it should act like a normal amplifier.  Adding the N-channel back in, you have a fourth inverting stage which is always "fighting" the P-channel.  Crazy that it works.  

Not quite.  Better to think of it as 2 inverters followed by a common-source n-channel circuit.  This gives three stages.  Output at drain inverts the input signal, so shunt-biasing from out-to-in works.  The p-channel is in parallel with the n-channel device.  It doesn't fight it, though.  An additional inverter between the second stage and the p-channel flips the signal.

I think that's right, anyway (the link doesn't work for me right now: overloaded the site's host?)

The higher than expected transconductance I mentioned in my initial post (over 5 bleedin' years ago) is due to most modern ICs being buffered (as shown in the data sheet I linked to).  Text-books/other data-sheets show a simplified, 'unbuffered' schematic, which is what I was expecting when I plumbed the circuit in originally.

[IGR]

Sure it is not guitar pedal but something completely different:
http://kambing.ui.ac.id/onnopurbo/orari-diklat/teknik/qrp/HIS%20TRX,%20A%20TWO%20BAND%20ONE%20CHIP%20ONE%20TRANSISTOR%20VXO%20CONTROLLED%201%20WAT
I thought it could be maybe helpful, gain stage remains the same.

[earthtonesaudio]

Ah, yep.  I stand corrected.  Thanks Gez!