THE ENGINEER'S THUMB... At last, a better compressor!

[Sanguinicus]

Hit this link for my voltages.

https://www.dropbox.com/s/adqnkn3yy2q888s/ET%20voltages.png?dl=0

Should have mentioned the Q also. C=4.5, B=2.5, E=4.0

All these are obviously effect engaged. No signal, just guitar plugged in. Needed to hold the dmm probes.

Vero build here:
https://tagboardeffects.blogspot.com/2012/11/engineers-thumb.html

[merlinb]

Should have mentioned the Q also. C=4.5, B=2.5, E=4.0

Something funny going on with the transistor and pin-5 of the CA3080. Collector & pin-5 should be around 0.6V, not 4.5V.  Also, you said B = 2.5V, but on the diagram you posted you've written IC2 pin-1 = 3.7V? They should be the same since they're connected together.
Looks like the problem in in that area, so check your connections there. Almost looks like C & pin-5 are not connected together, or the transistor is the wrong type / wrong way around / bad.

[Sanguinicus]

Yeah I just found an error on my board. Now I fixed that but now it sounds like an overdrive. The Ratio control increases gain and volume. Level works as per normal, still no noticeable treble control.

It's a pretty sweet overdrive to be honest, But I came here for compression!

EDIT: After doing extra reading, I found that by even measuring the bias input (pin 5) of the OTA, you can fry it. So perhaps this is what happened and why it's sounding like an overdrive. Time to hunt for a new one, or a LM13700 i think.

[changes]

Hello

I've built this LED indicator mod and works pretty good but I am interested in the opposite indication. Dark when compressing... Any help is more than welcome

> the ouput of that opamp to B of a NPN. Simple enough.

That won't work. The LED is always-on and carrying infinite current. Also fouls-up the compressor.

There's a simple linear-proportional way to add an indicator. See the BC327 PNP and 1K resistor? This flows a current proportional to over-level, and closely related to amount of gain reduction. That current is weak and already committed to tickling the LM13700. But we can take the same Base signal, to a second PNP, with a smaller resistor, and get a proportionally larger current we can dump to an LED.

(This form of "current mirror" has a small 40mV mirror-error which is totally moot for a visual indicator.)



The light happens when the compressor is "doing something", which makes sense to me. If the LED is off, it is just a buffer/booster. When signal gets loud, so that the sidechain tickles the '13700 to turn-down the signal, the LED turns-on and gets brighter.

For bench-test only, you can put your volt-meter *across* that 1K resistor. Use a 2V range (auto-ranging may be confused by dynamic guitar signals). The DC voltage is zero for no- or small-signal. When signal is excessive, voltage will increase to 1V (maybe a bit more or less), fading back to zero as the note decays.

[PRR]

> Dark when compressing...

Dark-emitting diode.

[changes]

 

Wish there was one.

> Dark when compressing...

Dark-emitting diode.

[PRR]

[merlinb]

I've built this LED indicator mod and works pretty good but I am interested in the opposite indication. Dark when compressing... Any help is more than welcome

This *might* work if you play with the resistor value, but LEDs don't vary their brightness in a convenient way so it might not be possible to get a useful variation in brightness with signal level. Easy enough to try though. Try 1k to 2k to start with.

[changes]

I've built this LED indicator mod and works pretty good but I am interested in the opposite indication. Dark when compressing... Any help is more than welcome

This *might* work if you play with the resistor value, but LEDs don't vary their brightness in a convenient way so it might not be possible to get a useful variation in brightness with signal level. Easy enough to try though. Try 1k to 2k to start with.

Thanks guys

I'll try both asap and get back with the results

Cheers

[jonny.reckless]

I've recently had a few people message me about PCBs for the "Engineer's thumb redux" I did a couple of years ago. This was based on the ET with JFET input and the inclusion of a noise gate into the compressor. I don't have any boards left, but you can find the Gerbers and NC drill files for the board I made, along with the schematic and bill of materials in my shared Google drive folder:
https://drive.google.com/drive/folders/1usNiADluT3tfycfkl7Y7Yo24fzZkVLZN?usp=sharing
Feel free to make some more boards from these Gerbers if you like.

[rankot]

Hi Merlin, I'm referring to your issue 4, where you use LM13700's buffer for output. Would paralleling those transistors give less noise? Like this, for example:

[merlinb]

Hi Merlin, I'm referring to your issue 4, where you use LM13700's buffer for output. Would paralleling those transistors give less noise? Like this, for example:

I doubt it would make the slightest difference since most of the noise comes from the OTA itself, but you can try. I'm not sure how well matched they'll be; they're on the same die but they're also Darlingtons. I don't think it will hurt anything though.

[PRR]

If a competent designer (say, Merlin) put an emitter follower *after* a volume control, then noise should be a non-issue.

I don't see how it would help or hurt, or be worth the effort to change an existing PCB/layout.

[rutgerv]

After working my way through the wonderful pages of this thread, I feel enthusiastic about building my own ET. I have plans to build it into a guitar box that I plan to use for silent guitar recording. I designed a variable impedance buffer (just for the heck of it), some tone controls (bright switch, mid scoop, high-pass and low-pass filters + wah), and was considering a JFET based pre-amp circuit before going into either (a) my computer or (b) a stand-alone power-amp + cab simulator.

I was planning to put a compressor before the pre-amp. Because I come from designing synths I have pile of modern SSI2164 VCA's, which I use for the tone control section and wah as well. I was wondering if it would be worthwhile (in terms of THD+N performance) to redesign the ET for using this part (instead of the LM13700). What do you think?

I recall from the datasheet that 2164's have their best performance around unity gain, which means the 'OTA inside feedback-loop' design is probably no longer needed. Also, 2164's can be paralleled to achieve even better performance (if needed). One drawback: the control law of input of the 2164 is not linear (like the LM13700) but logarithmic, so the current detector+envelope circuit can't be connected directly. One way to overcome this is to use a second section of the 2164's to linearise the first section? Or would there be alternative detector+envelope approaches that put out a dB-linear control signal?

Regards,

Rutger

[jhsa]

I finished my latest batch of changes to the gate design. I added a new peak detector, and a threshold comparator with some hysteresis. I also added separate attack and release controls for the noise gate. It's now an 8 knob monster

Here is the schematic:


Here are some photos of the revision C PCB:





Here it is in action. Sorry for the poor audio quality, it's just my phone camera, which is adding some weird compression of its own to the sound
One day I'll get a proper camera and mic for this stuff.

Overall I am fairly happy with the circuit design. It's a bit more complex than I would have liked, and at 100 x 70mm will no longer fit inside a 1590B (you need a 1590BB instead), but the added flexibility to the noise gate is nice. Let me know if you would like a PCB to build yourself.

I know it's been a while since this was posted
Someone just pointed this version to me. I did build the original ET with all flavors and I love it, but I am probably going to give this one a try as well I have had a quick look at the schematic and noticed a couple things. No input or output caps. Normally pop sounds when switching are caused by DC. Caps will block DC.
Also connect pin 1 of the MAX1044 to pin 8 to increase the operating frequency to something well above the audio spectrum. If using the ICL7660 instead, which is what I have, use the 7660S and also connect pin 1 to pin 8. I believe you could then even remove TR3 and it would still be silent
Can't wait to give this a try.

[dbarrie]

So I've built up merlinb's version 2 of ET with the modifications mentioned to make it suitable for bass (namely, the 220k release resistor and 2.2uF caps on the input to opamp C) on a breadboard, but as a few other people mentioned in this thread, I'm getting some awful distortion when using the compressed signal. All the voltages I've measured seem reasonably close to what others have shown is correct.

Some details about the build and what I've done so far to debug it:

The build is sitting on a breadboard right now. I've got it laid out pretty clean, with the component leads cut short and hook-up wire connecting things that need it. Obviously a breadboard is going to cause some issues, but the amount of distortion I'm hearing seems like way more than can be attributed to the stray capacitance of the breadboard! I've got the pots and jacks hooked up using additional lengths of hook-up wire. The circuit is powered by a 9V wall-wart that measures at 9.22V. I didn't have any TL072s or TL074s on hand when putting it together, so I've used 3 TL071s as the A, B, and C opamps, and an LM358 to generate the 4.5V rail. I've not connected any of the offset pins of the TL071s - they're just floating. I'm also using an LM13700.

My bass has an on-board EQ (made of 3 TL072s, bass/mid/treble). I get an output of ~100mVpp with it disabled, and ~200mVpp with it enabled and zeroed out.

When probing the circuit, I see the base of the transistor oscillating between ~4.2 and ~4.75V at 60Hz, obviously picking up some hum from the computer, monitors, power supply, etc. When playing the bass, I do see it immediately turn into a very narrow pulse and then split into two (one smaller than the other) as the signal dies down. Adjusting the attack pot changes the way this behaves, and it feels like that part is doing something that seems maybe correct, if you'll pardon my inability to describe it very well. If anyone can describe what that signal should look like, I'd love to be able to compare it to what I'm seeing.

The compression part does definitely appear to be working. When I let a note ring out, I hear it start out loud and quickly get compressed as the attack kicks in, followed by it being amplified to around the same level where it would normally start fading away. Changing the attack knob can give me a bit of a pulsing or "wooshing" effect. It just sounds really, really distorted and bad!

Is there anything obvious I should try? Is that 60Hz noise on the transistor's base the likely culprit here? What should that signal look like when not playing? Are there other points I should try probing?

I can post oscilloscope screenshots if people would find them useful!

[jonny.reckless]

I know it's been a while since this was posted
Someone just pointed this version to me. I did build the original ET with all flavors and I love it, but I am probably going to give this one a try as well I have had a quick look at the schematic and noticed a couple things. No input or output caps. Normally pop sounds when switching are caused by DC. Caps will block DC.
Also connect pin 1 of the MAX1044 to pin 8 to increase the operating frequency to something well above the audio spectrum. If using the ICL7660 instead, which is what I have, use the 7660S and also connect pin 1 to pin 8. I believe you could then even remove TR3 and it would still be silent
Can't wait to give this a try.

Let me know if you want me to share the Gerbers or NC drill file for the 8 knob redux version.

[Rob Strand]

Let me know if you want me to share the Gerbers or NC drill file for the 8 knob redux version.

I like your London bridge PCB holder.

[jonny.reckless]

I like your London bridge PCB holder.

https://www.amazon.com/Aven-17010-Adjustable-Circuit-Holder/dp/B00Q2TTQEE/ref=sr_1_2?dchild=1&keywords=circuit+board+holder&qid=1596515622&sr=8-2
This was bought for the purpose unlike my cable organizer which is a hat rack

[merlinb]

Let me know if you want me to share the Gerbers or NC drill file for the 8 knob redux version.

I like your London bridge PCB holder.

Tower bridge