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

[snk]

Excellent !
Thank you, Kevin !

[tubegeek]

Add one more tested 5-knob to the roster. (It's not boxed up yet, but it sounds great!)

Sweeeeeeet unit!

I built mine from Harald's (sabrotone) vero layout with these changes: I put 220R in series with the connection from Q1 collector to pin 1 of the LM13700, and added another connection from Q1 to pin 16 via another 220R. Then I connected 15 to 2, 14 to 3, 13 to 4 and 12 to 5, i.e., I used both OTAs in parallel. I also added .1 ceramics from V+ to V- on each chip - these are not provided on the vero layout. (On the two op amps, the cap fits OK under the IC socket. On the OTA chip, I put the bypass cap on the copper side of the vero.) My Q1 is a 2N3904, reversed with respect to the pinout of the BC327.

Note that, had I noticed that the layout only used one OTA before I started, I could have accomplished the paralleling just by omitting several strip cuts (column 20/rows d, e, f, and g.) I crammed the 220R's in by cutting the crosspieces away from my IC socket - had I done this from the get-go, I could probably have easily fit them underneath with the socket intact. I've attached a couple of post-op pictures of the 16 pin socket, front and back. It would have been MUCH neater to just omit the strip cuts in the first place, of course.

I haven't yet added the yin/yang input clipper LEDs, but I expect to piggyback them between E1 and D1 by soldering them to each other first and then taking only two legs through those holes.

Works great - very clean neutral action, but nothing "dead" or "clinical" about it - really hits the sweet spot. The output level is about 5V pk-pk before clipping which is an awful lot for a guitar pedal! Truly an ace design by merlinb - thank you for sharing this cool effect with us, no wonder it's such a popular build. I'll be drilling a box for this tomorrow!

[merlinb]

Truly an ace design by merlinb - thank you for sharing this cool effect with us, no wonder it's such a popular build. I'll be drilling a box for this tomorrow!

Thanks I'm glad you like it! I'll be posting an update to the ET soon which includes a compression indicator and output buffer, plus a minimalist version that only uses 2 opamps.

[tubegeek]

Thanks I'm glad you like it!

What's not to like?

I'll be posting an update to the ET soon which includes a compression indicator and output buffer, plus a minimalist version that only uses 2 opamps.

A maximalist option: taking off from here: https://www.diystompboxes.com/smfforum/index.php?topic=108570.0

I'm looking at a two-band implementation, sort of attempting to fit a Rane DC-24 into a pedal, with fewer (but still a lot of!) options. So far I think it'll look like this: Discrete JFET input buffer, PRR's clever crossover feeding two Threshold controls, an ET behind each Threshold, and an output buffer/virtual earth mixer driving the output.

Crossover discussed here: https://www.diystompboxes.com/smfforum/index.php?topic=112437.msg1037743#msg1037743

I've posted CamH132's drawing here which seems like it was derived from something PRR drew up the thread, now lost due to Tinypic being de-funked. Best I can do for PRR's drawing, is a version I recovered from Google image search in somewhat blurry form, but between the two drawings I think I get the idea:






A Lot Of Knobs for sure: I'm looking at which controls might best be suited to the nice dual concentric pots Smallbear stocks, which might be a nice way to simplify the front panel. So far I'm pretty sure the two Level pots should be done that way into the virtual earth mixer. And so far I'm pretty sure it DOESN'T make sense to put Attack, Release, and Ratio onto duals. That leaves Threshold as an open question.

Thoughts? Any bass players care to chime in? Jon "midwayfair" P had some excellent input at the topic linked above, but I'm seduced by the neatness of having two OTAs on the LM13700, so I'm loath to build two different compression engines for the two bands.

[DaveyJW35]

Here's a picture of my Enginner's Thumb Redux built using a pcb from Jonny Wreckless. I used some beet up pieces of sheet metal, folded EHX style, to give it some mojo.

[jhergonz]

I just finished reading the whole 32pages of this thread, and I think I never waste even a single second of reading.

thank you Merlin for designing a great circuit, I'am planning to use the ET version 1 for my friend's built in compressor for bass amplifier. I will use both lm13700 in parallel. and will power it with -+15. I hope it will work.

[DrAlx]

I have built the version 4 circuit using my own layout.
With the level pot at minimum, I hear a weak thump at the output whenever I strum my guitar strings.
I never had that problem with the version 1 circuit.  It was dead quiet with the level at 0.

I think the problem is the output buffer, because if I audio probe directly on the middle pin of the level pot (or on pin 7 of the OTA) it is quiet but pin 8 thumps when I strum.

I am using an LM13600AN instead of an LM13700.  Is the version 4 circuit only suitable for the LM13700 ?
Looking at the datasheet the buffer in the LM13600 seems to be affected by the OTA control current, and I guess that is what is giving the thumps when I strum.

[slashandburn]

I have built the version 4 circuit using my own layout.
With the level pot at minimum, I hear a weak thump at the output whenever I strum my guitar strings.
I never had that problem with the version 1 circuit.  It was dead quiet with the level at 0.

I think the problem is the output buffer, because if I audio probe directly on the middle pin of the level pot (or on pin 7 of the OTA) it is quiet but pin 8 thumps when I strum.

I am using an LM13600AN instead of an LM13700.  Is the version 4 circuit only suitable for the LM13700 ?
Looking at the datasheet the buffer in the LM13600 seems to be affected by the OTA control current, and I guess that is what is giving the thumps when I strum.

I hate to be "that guy". There's probably a logical answer to this but why would you want the output level at 0?

[DrAlx]

I have built the version 4 circuit using my own layout.
With the level pot at minimum, I hear a weak thump at the output whenever I strum my guitar strings.
I never had that problem with the version 1 circuit.  It was dead quiet with the level at 0.

I think the problem is the output buffer, because if I audio probe directly on the middle pin of the level pot (or on pin 7 of the OTA) it is quiet but pin 8 thumps when I strum.

I am using an LM13600AN instead of an LM13700.  Is the version 4 circuit only suitable for the LM13700 ?
Looking at the datasheet the buffer in the LM13600 seems to be affected by the OTA control current, and I guess that is what is giving the thumps when I strum.

I hate to be "that guy". There's probably a logical answer to this but why would you want the output level at 0?

It's noticeable at low level too regardless of how the ratio pot is set. If you never turn the level down then the noise from the jumps in bias current are buried in the the guitar sound and so you don't notice them.

[merlinb]

I am using an LM13600AN instead of an LM13700.  Is the version 4 circuit only suitable for the LM13700 ?

Hmm that's could be it, I haven't tested it with an LM13600. Easy enough to eliminate the emitter follower and build it the old way.

[DrAlx]

Hmm that's could be it, I haven't tested it with an LM13600. Easy enough to eliminate the emitter follower and build it the old way.

My layout was for a PCB so I ordered an LM13700 rather than hack the board.  LM17000 solved the problem.

[jubal81]

Any benefit to using a VCA rather than an OTA?

[Kevin Mitchell]

Any benefit to using a VCA rather than an OTA?

What?

I thought these compressors were an envelope detector controlled VCA.

-KM

[PRR]

> using a VCA rather than an OTA?

Is there a difference?

(An OTA is a building-block often used with other parts to make a VCA.)

(Randomly changing one to the other will force other changes throughout.)

[merlinb]

Any benefit to using a VCA rather than an OTA?

[Ben N]

Any benefit to using a VCA rather than an OTA?

Amen! Preach it!

[Sanguinicus]

Been wanting to build this thing for ages. Finally did. But it's got a problem.

I've built the two-knob version for now. The level knob works as expected, max is basically unity volume (is that correct?).
The problem is with Ratio. Fully counter-clockwise (or fully clockwise, i get the pot pins mixed up) it basically sounds like the bypassed signal. Turn it up slightly and it goes through a very brief staticky/scratchy transition... to basically the same thing. Now with a bit of noise in the signal and possibly kind of, sort of compressed, there's slight background noise too. The rest of the sweep does nothing to affect the signal.

Treble boost doesn't seem to work either, could be unrelated. But I'm not fussed with getting that working at the moment.

Anything obvious I should check? I've done the whole remeasuring, reflowing, etc apart from voltages. I'm using the Tagboard Effects vero layout with CA3080 (known working, used it in a previous comp). I really want to get this working.

[merlinb]

Anything obvious I should check? I've done the whole remeasuring, reflowing, etc apart from voltages. I'm using the Tagboard Effects vero layout with CA3080 (known working, used it in a previous comp). I really want to get this working.

Post your voltages, and a link to the layout you used.

[Sanguinicus]

Anything obvious I should check? I've done the whole remeasuring, reflowing, etc apart from voltages. I'm using the Tagboard Effects vero layout with CA3080 (known working, used it in a previous comp). I really want to get this working.

Post your voltages, and a link to the layout you used.

is there a reference of what the voltages should be? So I can check as I go.

[merlinb]

is there a reference of what the voltages should be? So I can check as I go.

For the layout you're using, I don't know, maybe you can contact whoever made it. People swap the opamp sections to suit their needs, and the CA3080 is another deviation from the stock circuit. There is no universal "this pin should always be X" formula. If there is a schematic to go with the layout you're using then it would be ideal if you wrote the voltages directly on it. Then we will probably be able to see at a glance what it wrong.
The voltage chart for my Iss.3 version may help: http://valvewizard.co.uk/engineersthumb3.pdf