easy ways to create compression within a boost/od design

[amonte]

I was thinking about this earlier today - I know there are compressor schematics out there, but are there easy ways to create a little bit of compression within a design?

[Joe Hart]

If you boost the signal way up, then chop it down (like with diodes to ground), it will, at least, create a limiting effect.  I think....?
-Joe Hart

[strungout]

Oy.

I was gonna try this sometime: using an ldr (in series with a resistor), in the feedback loop of an opamp, that would be heatshrunk together with two (or one) LEDs either in the feedback loop also or shunt to ground to modulate it. Those LEDs light up following the signal strength, at least when the gain of the stage driving it is high enough. But maybe it would be too "rough" of a kind of compression...

Edit: Erhm, I think there's already compressors that use this method?

[petemoore]

Flatline sounds like what you're describing...
 The signal lights the LED which causes the LDR to impart a limiting effect on the amplification  [I think by altering negative feedback, been a while since I looked at it].

[Mark Hammer]

There was a clever design in Electronic Musician in the mid-80's called the "Fuzztain".  The author used a pair of CLM6000's in a clever way.  The LEDs served as the clipping elements in a pretty standard back-to-back arrangement in the feedback loop of an op-amp.  Of course, since the LEDs cause LDRs to change resistance, the LDRs were placed in the feedback loop of another op-amp at the input of the pedal.  

Slam the strings and you get a momentary clipping from the LEDs, at which point the LDRs lower resistance and the signal amplitude reaching the LEDs is reduced.  Of course in this application the amount of compression produced affects how much distortion is produced since anything that lowers signal level reduces the likelihood of the signal being above clipping threshold downstream.  What was not attempted was placing the *compression* downstream, such that full dynamics were retained for the signal hitting the clipping stage but restricted dynamics imposed post-clipping.

Alternatively, since there are two LDRs at one's disposal in such a design, one LDRs gets used for lightly compressing the signal upstream (pre-clipping) and the other one gets used for similar light re-compression downstream (post-clipping).  Could be a useful alternative.

The other thing that was never exploited is to use the LDRs for a swept filter.  This presents interesting possibilities because swept filters *become* more interesting sounding the richer the harmonic content you feed them.  In this case, the sweep of the LDRs (and whatever filter it is part of) is increased when the signal is pushing harder against the LEDs (more clipping).

[David]

Oy.

I was gonna try this sometime: using an ldr (in series with a resistor), in the feedback loop of an opamp, that would be heatshrunk together with two (or one) LEDs either in the feedback loop also or shunt to ground to modulate it. Those LEDs light up following the signal strength, at least when the gain of the stage driving it is high enough. But maybe it would be too "rough" of a kind of compression...

Edit: Erhm, I think there's already compressors that use this method?

The Flatline does this exact thing.  It's smooth as silk!

[petemoore]

ORange Clipper...anyone try an OS with diodes to Gnd at the output?

[Aharon]

There's another thread regarding the H11F1 opto and maybe this is something where this element could be used.
http://www.fairchildsemi.com/ds/H1/H11F1.pdf
Aharon

[Mark Hammer]

ORange Clipper...anyone try an OS with diodes to Gnd at the output?

Build the modded version I suggested (posted at GGG on the OS page), and stick a couple of diodes in the feedback loop of the gain recovery stage, making sure to up the gain a little in that stage.