Time for another annotated stripboard, this time Sam’s Simple DIY Envelope Generator:
The trigger to gate stage confused me a bit; see discussion here.
10 Likes
Much to my shame but I’ll ask a stupid question.
Does the ground from the jack sockets connect to the ground of the power supply?
Built this circuit, on power up the supply gets very hot quite quickly.
3 Likes
All grounds should be connected to each other and to the ground (0 V) from the supply. You want all parts of the system to agree on where 0 V is.
If your supply gets hot with only this module plugged in something’s wrong; this uses very little current.
5 Likes
This will absolutely save me the next time i build this. I had to fiddle about with it a bit.
3 Likes
I’m a bit late to the party on this one but quick question: is the 10nf capacitor electrolytic, poly, ceramic?
Thanks!
2 Likes
Ceramic works, but maybe film are better ?
1 Like
You’d have a bit of trouble finding a 10 nF electrolytic. Mostly they tend to be above 100 nF.
I’d probably use film, but as @Dud says, ceramic should be good.
1 Like
Thanks both. I had thought that it would be small for an electrolytic but only the 1uf had been called out specifically as non-polarised. Appreciate the help 
1 Like
3 Likes
Based on a conversation in a separate thread with @analogoutput, it might be worth adding in a pulldown resistor on the output. The output sits at around -2v for me without one, but adding in a resistor to ground (I used 1M, but 100K suggested too) keeps it sat at 0v until the envelope is triggered.
1 Like
As mentioned here, R9 and D7 after R8 creates a voltage divider that reduces the maximum amplitude at the output to something like 6 V, which may not be enough to fully open some VCAs or VCFs. I think most Eurorack EGs put out something more like 8 to 10 V. If you move R9 and D7 to before R8 then the maximum output should be in something more like the 9 V range.
1 Like
The LED if positioned after the 1k output resistor also alters the shape of the envelope as seen in this simulation. Blue is with the LED after the output resistor and green is with the LED before it (with a voltage divider added to scale it to about the same amplitude for comparison). There’s also a 100k pulldown added before the output resistor in both cases.
3 Likes
i build this hoping to be able to trig it with a square lfo but it doesnt work.
why is that? (i was using the doepfer a 145)
for now i tryed only with pulses from turing machine or from a clock divider (trigged by a lfo) and it works with those kind of trigging…
I don’t know if you were using it in AR or AD mode, and I don’t know the amplitude of the LFO waveform, but let’s say you’re in AR mode and you have +5 V at the input jack. Then:
- There is a diode D1, and you drop around 0.6 V across that, so you’re at about 4.4 V
- There is another diode D3, with about 0.6 V drop, and there’s equal current through equal resistors R4 and R5 so equal drops across them, all adding to 4.4 V, so the drop across R5 is (4.4-0.6)/2 = 1.9 V
- This equals the voltage at the op amp + input.
Meanwhile the threshold voltage at the - input is 12 * (10/(10+47)) = 2.1 V.
In that case the input voltage isn’t enough to bring the + input above the threshold voltage so the envelope generator doesn’t trigger.
To me it’s not at all clear why there are two diodes in the path — D3 could be moved to the AD path, I think — or why there’s a 50% voltage divider (R4, R5), or why the threshold voltage is so high. Decreasing R4 or R7, or increasing R6, ought to fix it. I don’t see any reason the threshold couldn’t be a lot lower — R7 could be 4.7k or 3k or 2k or maybe even 1k and I wouldn’t expect that to be too low.
1 Like
according to the manual the doepfer 145 has an amplitude of ± 2.5 V
i finally would like to use this envelope with the triple lfo by soundbender:
yet i don’t know how much voltage spits out …
do you think is better trying just one of those suggestions of yours at time or all of them at the same time? couple of them? any risk?
I mean decreasing R4 + decreasing R7 + increasing R6
and moving d3 to …where exactly? after the switch before C1 and R1?
You could try all the possibilities on a breadboard, but if you don’t want to do that then lowering R7 (or raising R6) seems to me the most sensible. If it were me I’d probably just drop R7 to about 2k.
Changing R4 would also change the height of the envelope output and that might cause trouble. As for R6 it’s the ratio of R6 to R7 that matters so not much point in changing both. Moving D3 would help for AR mode but not for AD, and I think the threshold mismatch is even worse for AD mode.
i dont have a breadboard 
maybe i should get one of course…
i found on soundbender website this:
what. do you think?
only problem for me would me to buy another capacitor, here they use a 4uF and i have only 1uF non-polarized ones… unless i put in parallel 4 of them but is not that great in term of design and space …
seems a bit more rare or more expensive a 4uF non polarized capacitor…
are those ok?
would 4.7uF fit the same?
4.7 uF film
As it says here:
Capacitors tend to be lower precision than resistors. ±20% is not uncommon, and you sometimes even find ones with -20%, +80% tolerance — meaning that if it’s nominally 10 nF, it won’t be less than 8 nF, but it might be as high as 18 nF. Correspondingly, circuits usually aren’t designed to require capacitors to be very precise. If the instructions don’t say otherwise, a 20% capacitor will probably work fine. (In fact if you don’t have the right value cap but you do have one that’s within a factor of 2, there’s a good chance you can substitute it with little or no ill effect.)
The only thing really affected by the value of the capacitor here is how fast or slow an envelope you can get. 4 µF would allow envelopes 4x slower than the 1 µF of the original circuit. If you don’t need such slow envelopes you can continue using 1 µF.
Film capacitors above 1 µF tend to be large and expensive. Look for non-polarized electrolytics. Tayda has 4.7 µF for 12 cents:
(Or it’s possible to use two 10 µF back to back in series:)
1 Like
i entered in the gate a fast lfo to make sound and use the audio probe to check where the signal fail
it was working for a while then after trying removing diode from switch it started to be glitchy, i put it back and now is not working anymore
update:
with audio probe actually it’s at the pin 7 of the TL072 that sound stops…
there is sound at leg 5
i think ill trash everything and build another envelope. this doesnt work for me…
tryed with 4 non-polarized capacitors in parallel and tryed with 2 polarized electrolitic capacitors in series.
tryed resolder wires, checking for short etc…
i found out that by pulling one leg of “attack” potentiometer it was working… then it stopped…
i resolder that to another new pot… same… pulling working…resolder wire on the leg of the pot without touching any small hole nearby… works for a while then stops again… no matter i pull or not now