Thumb Sucker Compressor distorts instead

[matopotato]

Try a larger value for C6, like 2u2 or even 4u7, and see if that improves anything.

Ok, so I have original electolyte 1uF at C6. Instead of desoldering, I took another 1uF i parallel. I imagined it would be the same as a 2uF cap.
Interesting effect: Pop, LEDs flashed, then silence. And then gradually build up.
The tone was very bassy, so most of the tone gone. And the bassy still had the distortion. (Got some fuzz feeling sort of).
So I did not go with 4,7uF as I doubt it would improve.
Thanks for the suggestion regardless.

[matopotato]

Attempted to swap the C5 C6 from electrolyte to Film, but I feel te eyelets have just about had enough of my attempts.
So new plan is to breadboard with whatever spares I have.
Using the ICs, transistor and at least one pot.
Not sure what I will do with any result though. I guess I want to know if the kit will work for me. And how it sounds.

[anotherjim]

The OTA input is prone to clipping. R7 and R6 form a voltage divider to reduce the signal level to avoid clipping. Measure the values of those resistors to confirm they are correct. You could try increasing R7 above 10k to reduce the signal level the OTA has to pass.

[matopotato]

The OTA input is prone to clipping. R7 and R6 form a voltage divider to reduce the signal level to avoid clipping. Measure the values of those resistors to confirm they are correct. You could try increasing R7 above 10k to reduce the signal level the OTA has to pass.

Thanks. Wasn't aware of that.
Seems OTA in compressor has a few less desirable side effects.
R7 9,78k and R6 218R, so they check out. But if R7 could be increased if anything, then currently being below 10k is not helping I guess.

[puretube]

Tried a 100n cap (ceramic or film) or 2 or even 3 across the power-pins of the opamps/OTA yet? (In fact in parallel to C100, but closer to the IC-pins). And/or maybe 1 in parallel to C101?

Ok, so I tried 100n and also 220n across pin4 amd 8. No change or effect either way.
Also on C101 but no effect.
Was the expected result that the dist would go away, or be reduced?
Thanks for suggesting though, appreciate it.

Just to avoid any rubbish on the power-/vref-rails (be it from the circuit itself, or from outside) entering sensitive points in the circuit ... (despite the good PSRR of modern ICs).

[puretube]

Another thought: How about swapping the position of R8 & C4?
(i.e.: instead of the opamp directly driving a 10µ cap which is grounded through 11k,
having the opamp driving the cap, which is grounded through 10k, via 1k).
The Valve-Wizard did that in his Version 2, too.

Bingo?

[matopotato]

Another thought: How about swapping the position of R8 & C4?
(i.e.: instead of the opamp directly driving a 10µ cap which is grounded through 11k,
having the opamp driving the cap, which is grounded through 10k, via 1k).
The Valve-Wizard did that in his Version 2, too.

Bingo?

Thanks for the idea.
Right now I am breadboarding a copy but same IC and transistor plus one B1M pot. (Which ripped en eyelet, so original build is drifting slowly towards "don't know why" land)
Will try to test your suggestion somehow. If breadboarding still does distort, I will try this.

[matopotato]

The OTA input is prone to clipping. R7 and R6 form a voltage divider to reduce the signal level to avoid clipping. Measure the values of those resistors to confirm they are correct. You could try increasing R7 above 10k to reduce the signal level the OTA has to pass.

Like 12k or 15k? Or higher still?

[anotherjim]

If you want to try it on your PCB version, lift one end of the 10k and connect it in series with a pot. Tweak and see. If it works, take the pot out of circuit then measure it and add a resistor of the pot value or next value up.

I've a hunch that (if this is the problem) 22k instead of the 10k may be the charm, but swapping resistors to find perfection soon gets tedious so a 50k or 100k pot should give enough range to try.

[matopotato]

If you want to try it on your PCB version, lift one end of the 10k and connect it in series with a pot. Tweak and see. If it works, take the pot out of circuit then measure it and add a resistor of the pot value or next value up.

I've a hunch that (if this is the problem) 22k instead of the 10k may be the charm, but swapping resistors to find perfection soon gets tedious so a 50k or 100k pot should give enough range to try.

Thanks. Got to 15k but heard no difference. Will try with a trim after the 10k. But I no longer have very high hopes.

[matopotato]

Now I have breadboarded the Thumb Sucker using other units I had in my "stash".
Bottom line: Sounds same as before. https://soundcloud.com/user-558594533/thumbsuckerscrapy-1?utm_source=clipboard&utm_medium=text&utm_campaign=social_sharing Distorts still. Cannot dial it out.

I tried swapping one of the OpAmps for a spare one, and felt it made a difference, but after rotating all three around I would say it makes no difference.
For a bit it sounded clean and nice, but not much of compression. Turns out I had forgotten to connect a few pins on the OTA LM13700.
Then it went back to the same distorted. Even sounded like a Fuzz for a while.

I tried my favorite "improvements" by using jrc 4562 for OpAmps, 18V for more headroom thus handling more of "distortion" before it breaks through, PNP with half and double hfe values.
Nothing made a difference this time. Sadly, since that was an alternative earlier.

I tried the swap of R8 and C4 suggested by @puretube, but sadly it had no effect.

I also tried R7 at 12k and 15k as suggested by @anotherjim, and still no change.

I have had it tried 200 miles away, so I doubt my house, room etc play a big role in this.
I run it with Strat Single Coils to Marshall DSL40CR. Have tried with humbuckers as well, but same bad.

I did "reuse":
the 3 ICs, and swapped OpAmps around as mentioned.
the PNP transistor and swapped around.
the Threshold B1M pot, but put it as the Release instead, so if that would have been the culprit I would have expected a different bad when in a different position.

So the only thing on the breadboard that is equal to the kit build is the OTA, LM13700.
On the other hand I tried another LM13700 in the build last weekend (the rest being the same) and it did not change anything, so I ruled it out.

To summarize, I feel I have swapped everything at least once, and in my humble opinion something should have changed...
I am very happy for all suggestions, but only one did some change, but to the worse sort of, although it was kind of fun (2uF for C6).
I thought breadboarding would make it fly and play properly. I would then have de-soldered and replaced one by one the components on the breadboard with one from the original kit until the offender revealed itself.
Then get a new pcb and re-build from the breadboard.
Or get a new kit with everything, but from today's tests I do not feel confident that it surely would work. And then I would have wasted another $50-ish.

I build a Zirconia from Lectric-FX, an optical compressor, and that is very nice. A certain smoothness to it as @Mark Hammer mentioned.
I was hoping to build something similar to Keeley Compressor/MXR Dynacomp/Ross compressor and felt the Thumb Sucker would fit that bill.
Anyone with some alternative build to the Thumb Sucker but that would be like Keeley preferably?

[matopotato]

If you want to try it on your PCB version, lift one end of the 10k and connect it in series with a pot. Tweak and see. If it works, take the pot out of circuit then measure it and add a resistor of the pot value or next value up.

I've a hunch that (if this is the problem) 22k instead of the 10k may be the charm, but swapping resistors to find perfection soon gets tedious so a 50k or 100k pot should give enough range to try.

Thanks, it was an interesting thing to try.
I put the 10k back on the board an added a 50k trimmer together with it. So I could dial 9.7k to 59k (ish). Sadly there was no change in the distorted sound. It gave an overall volume boost though.
Even though I have tried similar idea in other builds, it just did not strike me until you suggested it. I hope to learn by now using trimmers and pots in areas where a changed R could make a difference.

[eh la bas ma]

When it comes to compressors that use a half-wave rectifier, the momentary fluctuations in compensating gain will show up if the decay of the rectifier is set too fast or too slow.  Indeed, the audible consequences of half-wave rectification in noise-gates, compressors, and auto-wahs is regularly reported by people as "distortion".

The Thumb-Sucker is not a bad design, but its use of an OTA driven by a half-wave rectifier makes the instantaneous response of the OTA susceptible to envelope ripple.  That's one of the good things about using LDRs in compressors and auto-wahs: they "smooth out the glitches" arising from half-wave rectifiers...at least if you let them.

The solution is usually either shortening the gain recovery time, so that one leaves ripple behind quickly, or lengthening gain-recovery time so that the circuit behaves like a slow-to-respond LDR.

So If i understand correctly, this contradicts what ElectricDruid told us ?

"You should be able to use a circuit with the controls in any position without something sounding terrible. Ok, certain settings are better than others, and extreme settings might not be that "musical", but the circuit should continue to function correctly. If  a compressor only works with the knobs within a certain range, either there's something wrong with it or the design isn't much good, in my view. So, yes, I think there's an issue."

Edit : From project documentation : "Release- The time it takes for the gain to recover after the loud sound has passed."

Looks like Release is the "gain-recovery time" control.

So, according to Mark, the Release control should be the key to stop the circuit from distorting the signal ? However, only Ratio or Threshold can do that, Release set anywhere CW doesn't change anything about the distortion, and CCW it makes it worse, at least on my build...

[puretube]

Last attempt: take out the two red LEDs near the input, or increase that input-series-resistor from 1k to 4k7 or 10k. Voila!

[matopotato]

Last attempt: take out the two red LEDs near the input, or increase that input-series-resistor from 1k to 4k7 or 10k. Voila!

Thanks.
Ok, will try the increase. I already tried pulling the LEDs. No change.

[matopotato]

When it comes to compressors that use a half-wave rectifier, the momentary fluctuations in compensating gain will show up if the decay of the rectifier is set too fast or too slow.  Indeed, the audible consequences of half-wave rectification in noise-gates, compressors, and auto-wahs is regularly reported by people as "distortion".

The Thumb-Sucker is not a bad design, but its use of an OTA driven by a half-wave rectifier makes the instantaneous response of the OTA susceptible to envelope ripple.  That's one of the good things about using LDRs in compressors and auto-wahs: they "smooth out the glitches" arising from half-wave rectifiers...at least if you let them.

The solution is usually either shortening the gain recovery time, so that one leaves ripple behind quickly, or lengthening gain-recovery time so that the circuit behaves like a slow-to-respond LDR.

So If i understand correctly, this contradicts what ElectricDruid told us ?

"You should be able to use a circuit with the controls in any position without something sounding terrible. Ok, certain settings are better than others, and extreme settings might not be that "musical", but the circuit should continue to function correctly. If  a compressor only works with the knobs within a certain range, either there's something wrong with it or the design isn't much good, in my view. So, yes, I think there's an issue."

Edit : From project documentation : "Release- The time it takes for the gain to recover after the loud sound has passed."

Looks like Release is the "gain-recovery time" control.

So, according to Mark, the Release control should be the key to stop the circuit from distorting the signal ? However, only Ratio or Threshold can do that, Release set anywhere CW doesn't change anything about the distortion, and CCW it makes it worse, at least on my build...

Thanks,
I'll leave it to the ones you mention to comment on contents.
My observation of the distortion is that turning knobs had no or little impact. Of course zero Ratio made the distortion almost go away, but so did any compression. And volume close to zero, but again, not useful.

[puretube]

Last attempt: take out the two red LEDs near the input, or increase that input-series-resistor from 1k to 4k7 or 10k. Voila!

Thanks.
Ok, will try the increase. I already tried pulling the LEDs. No change.

Okay - I give up  (until I`ll try a build, one day ... - with 30 years old NOS chips).

[eh la bas ma]

I just tried. No luck...

First I changed R1 1k to 10k, same. Then I desoldered both leds : still distorting, no change.

Matopotato, Wook22, i am curious to know if you listened to the sound clip I posted ? Do you have a similar distortion with the settings shown on the picture, or do we have a different symptom ?

https://soundcloud.com/ehlabas-ma-chaussure/engineersthumb-distortingwav-1

[matopotato]

I just tried. No luck...

First I changed R1 1k to 10k, same. Then I desoldered both leds : still distorting, no change.

Matopotato, Wook22, i am curious to know if you listened to the sound clip I posted ? Do you have a similar distortion with the settings shown on the picture, or do we have a different symptom ?

https://soundcloud.com/ehlabas-ma-chaussure/engineersthumb-distortingwav-1

Sort of, I think so. Check link in #30 above.
Some have mentioned earlier that if you go full on Ratio and almost zero on Threshold then there will be distortion.
But for me it happens a lot earlier.
A bit discouraging that you have the same issue. Then little or no point getting another kit.

Edit: did you have the settimgs as in the soundclip picture throughout?
Mine might be more obvious on the bass strings, or full chord.

[matopotato]

Last attempt: take out the two red LEDs near the input, or increase that input-series-resistor from 1k to 4k7 or 10k. Voila!

Thanks.
Ok, will try the increase. I already tried pulling the LEDs. No change.

Okay - I give up  (until I`ll try a build, one day ... - with 30 years old NOS chips).

Sorry. Still I really appreciate very much your suggestions!
Will try the R increase since it is breadboarded, but as @eh la bas ma tried it below, it might not help.