Thumb Sucker Compressor distorts instead

[Rob Strand]

Sorry if I'm not being clear - the reports of crackling units are not very common

as I pointed out years back I noticed the "potential" for misbehavior in the design ...
and decided to talk about it after reported incidents

obviously it's not an issue in everyone's builds

Thanks.  That explains a lot.

I recreated the problem by shifting the THRESHOLD control to another location in the side-chain
and when the transfer gain in that section gets high enough I get the said crackle
nothing else changed

by adapting a different FW rectifier circuit and doing more like the Signetics/That guys do
that problem no longer shows up ...

I could be wrong but it seems to me that THUMBS is set close to the edge of the safety zone
for most people the circuit operates without that issue creeping in

bottom line, like I've said many times - there's much to explore close to what v1 was about including rethinking the threshold thing ...

Very interesting it's right on the cusp.

In recent times I've been more critical of circuits using TL07x's because of the phase-inversion issue when the input swings too far negative.   When I look at the Engineer's thumb I see IC-C potentially going into phase inversion when the rectifier swing goes too far negative.  (That's going to stop the gain reduction working!)

There's quite a few pedals and bass amps which add diodes to help reduce the problem.  For example the input buffer on the Boss CH-1 Chorus,



Unfortunately that trick won't work with IC-C.  You would need to flip the polarity of the rectifier but then it output of the current source Q1 has the wrong polarity for the OTA.  Some re-engineering is required!

[Rob Strand]

It's been a while now, but as I remember, it was quite clear it was unwanted distortion. And it appeared with settings that would make my build quite limited as far as pot interval usage (for want of a better word). I would probably have been content if some extreme settings caused the distortion, but I felt that several of the pots were reduced to 25% of the scale if to avoid it by "dialing it out". So I figured there was something else in my build that was causing the main part of the problem.

I might have a quick poke around based on Eb7+9's comments.

[Rob Strand]

This hack fix might help put it on the right side of the distortion cusp.  Maybe try 1.2M first.



Better would be to re-jig the bias voltage of the rectifier and current source to sit at a higher voltage and leave the audio at 4.5V or a tad above.

Unfortunately either of these mods is only trying to step around the issue but the resistor change is easy to try on the existing PCB's/builds.