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

[micmac1]

It says that I should read 3.9V on pin 14 of U1 (TL074). I do measure 3.9V, but after a few seconds it goes up to a little bit over 8V.

Possibly that is only happening because your voltmeter is attached, allowing current to leak through the transistor while the opamp is desperately trying to turn it off.

Hello merlin, thanks for the reply!

In the meantime I got myself a better multimeter (UT61E+) with 10M input impedance. Wanted to buy one anyway and this was as good an occasion as any. My old one had only 1M. But still, after four seconds it veers from 3.9V to ~8V

I have a PedalPCB Thumb Sucker as well and on its equivalent pin 7 I measure a steady 3.9V, it never goes higher. I think the Thumb Sucker is somewhere between ET version 2 and version 3.

So I'm still wondering what this is about, whether it's due to my assembling of the pedal or due to any changes between the different ET versions. Did you (or anybody else here) build a version 4 ET and measure a steady 3.9V on pin 14?

I still have 2 PCBs left, so I think I'll build this again and check if my 2nd attempt behaves differently.

Thanks!
Seb

[micmac1]

Hi all,

I finished my second ET4. Still same measurement on pin 14, it's 3.9V when compressing and then it goes up to a little over 8V when idle. I don't think I did anything wrong. I went with the build instructions provided, leaving out the compression LED and jumpering R23 and the LED. Seems to be how it is.

I'll leave it at that. The ET4s are identical to my older ET"2.5" pedal, except that the ET4s have two knobs less and the signal sounds maybe a hair "fresher" coming out of the ET4s. They all work like a boss

Posted some pics here: Link to musikding forum post

Have a great day!
Seb

[merlinb]

I finished my second ET4. Still same measurement on pin 14, it's 3.9V when compressing and then it goes up to a little over 8V when idle.

OK I just dug out my ET and checked. It does exactly the same thing you describe! And I've just noticed why: in earlier issues I included a diode in the feedback loop of that opamp, but I was pushed for space and removed it from Iss.4. It's not really a problem, but you could always add it if it bothers you.

[micmac1]

Thank you merlin for the confirmation!

I'm not bothered by this. I was bothered by this when finishing my first ET4, because the build PDF mentioned a steady 3.9V and I didn't see that on my build. So I thought I must have done something wrong.

Maybe, if you can find the time, can you please update the build PDF accordingly, so that builders don't start chasing a non-issue (like I did).

Thanks again! Have a great day!

Seb

[DrAlx]

BC327 has absolute max rating of -5V for Vbe. So even with pin 14 going high, the BE voltage appears to keep within limits for that part.
But I'm using 2N5087 as the PNP in my ETv4 build, and that has a lower max rating of -3V for Vbe. So the BE voltage limit of the 2N5087 is exceeded when my ET is idle.  My build has so far seemed to survive without that protection diode across the BE junction, but I don't know if the 2N5087 has been damaged as a result. I reckon the diode across the BE junction should probably be included as in the earlier ET versions.

[merlinb]

BC327 has absolute max rating of -5V for Vbe. So even with pin 14 going high, the BE voltage appears to keep within limits for that part.

The reverse Vbe rating I think is related to the Zener behaviour of the b-e junction, so the max quoted on the datasheet probably assumes a supply that could dump dangerous levels of current into it, which the ET cannot do. But yeah, the diode is a good fail safe.

[PRR]

BE voltage limit of the 2N5087 is exceeded when

Then it comes to the maximum current. Current is limited by the 1k resistor and available voltage, say 5mA. The gain junctions in small devices often stand 10mA(*), tho the 2N5087 does not say that.

While I think it is OK, it does seem like the diode is a Very Good Idea.

(*) Up to a point, breakdown is not fatal to the major transistor parameters. However even slight breakdown can spoil self-hiss and offset voltage parameters. in The Thumb, neither is critical.

[GGBB]

I finally built one of these (always late to the party) and it's wonderful! Thanks Merlin. I really love how it sounds - reminds me of my first compressor 30-something years ago - which I have missed ever since - a Boss CS-2 - although I'm sure I remember the Boss more favourably than it actually was.

Does the brightness rating of the LED make a difference to how soon it turns on, out of interest? Mine's a 10,500 mcd clear blue, ratings for super bright LEDs vary from a few thousand mcd to over 30,000 mcd.

Yes it will do. The brighter the LED, the more light you get for a given current, so the tiniest currents should become that bit more visible.

The only trouble I had was with the compression LED - it only barely lit up when strumming very hard - even with a super bright (supposedly 8500mcd) white LED. A plain red LED worked about as well. I ended up increasing the parallel resistor to 22k and now I have a nice response with either LED - I chose red though because it faded out a little less abruptly which looked better to me.

Could this change cause any problems? I'm not noticing anything obvious. Does it do anything like lower Vbias current and if so does that even matter?

I also added a brightness knob - I replaced the treble boost switch with a C50k pot using pins 2 and 3 and changed the 10k resistor to 4k7 (4n7 cap the same). I find this quite useful and a lot more versatile than the switch.

Definitely going on the pedalboard - replacing my CS-2 clone.

[merlinb]

I'm glad you like it! It remains my proudest pedal design.

The only trouble I had was with the compression LED. I ended up increasing the parallel resistor to 22k and now I have a nice response with either LED
Could this change cause any problems?

My instinct is to say yes that's fine. If your circuit isn't horribly distorting then I guess there's your proof! I am surprised your white LED was dim though, but then again, LEDs are annoyingly variable.

[GGBB]

My instinct is to say yes that's fine. If your circuit isn't horribly distorting then I guess there's your proof! I am surprised your white LED was dim though, but then again, LEDs are annoyingly variable.

Thanks Merlin. I was surprised about the LED too - the white one is extremely bright normally. And the red one which worked just as well is supposedly just 600mcd compared to the white's 8500mcd. Seems wrong. Maybe Vf is part of the issue - the whites are ~2.7V whereas the reds are ~1.6V.

[patricks]

The only trouble I had was with the compression LED - it only barely lit up when strumming very hard - even with a super bright (supposedly 8500mcd) white LED. A plain red LED worked about as well. I ended up increasing the parallel resistor to 22k and now I have a nice response with either LED - I chose red though because it faded out a little less abruptly which looked better to me.

I'm going to try this, thanks!
I love mine, I stopped fiddling with the LED after trying two or three different colours. It lights up on mine but only when it's limiting, and then your ears tell you what's going on, you don't need an LED 

It did take me a little while to get used to just how versatile the controls are, but I think I've got the hang of the basics