NN Simple Comp.... is it any good?

[iainpunk]

http://experimentalistsanonymous.com/diy/Schematics/Compressors%20Gates%20and%20Limiters/NN%20Simple%20Comp.gif

has anyone tried this compressor schematic? looks good to me, but im away from my hobbyshack for a while. im looking to add a compressor at the end of a ringmod/octave/fuzz pedal because its volume is very uneven (envelope dependent) and i have a single unused opamp, since i use a quad opamp.

noise only matters a little bit, since controlled noise is the purpose of the pedal, but linearity and sound are of less importance because it is going to be followed by an active tonestack.

[Mark Hammer]

That's very close to the circuit Roland used for the "soft distortion" part of the Funny Cat.  That is, a gain stage otuput is rectified and drives the gate of a FET that is used to determine the gain.

The extent to which such a circuit results in smooth compression versus soft distortion (which was essentially audio-frequency envelope ripple) will depend on the smoothness of the rectification, the relevant time constants, and how much the FET changes the gain.  A resistance to ground in parallel with the JFET might help, although the variable gain control may do all that's required to keep the circuit in line.

[iainpunk]

breadboarded it today, it is more of a novelty/cool circuit than a practical one.
don´t think its a good compressor for my purposes, its more of an overdrive indeed.

with a little modification the pedal im working on has even volume independent of input volume, which is a nice thing to have.
i still have a leftover opamp, Any ideas what to do with it??

[duck_arse]

did you use a BF256C or something with slightly tighter specs?

[deadastronaut]

checkout sams comp...

https://www.diystompboxes.com/smfforum/index.php?topic=101118.msg1125364#msg1125364

[Mark Hammer]

The Univox Uni-Comp (licensed to other companies) was my first compressor and remains my favorite, after going through a dozen or more over the last 40 years.  It is dead simple, if a little unorthodox, but has all the things one likes about optical compressors.  NOT "transparent".  I use it as an effect.  But fairly quiet.  Weird about that diode pair on the output.

[PRR]

> Weird about that diode pair on the output.

The detector is also diode-based so if the gain-control were perfect the output diodes would never conduct.

The LDR gain control is slow, so initial transients may be BIG, and may splatt the next stage. Clean clipping may be preferable.

We find this even in broadcast quality limiters. Attack time is never zero, and clippers are used to keep control for the first milliSeconds so the transmitter doesn't over-modulate.

[Mark Hammer]

Interesting!  So because the diodes are only conducting for a few milliseconds, you don't hear what they're doing as clipping/distortion.  They're just taking care of things until the rectifier and photocell can get out of bed and get dressed.  Clever.

And here I thought I liked it because it was simple and cheap.  Never realized it had a trick up its sleeve.  I may be remembering wrong, but I believe I used Schottky type for the rectifier, and silicons for the pair on the output.