Hi - I recently built a dual envelope generator built on the super simple envelope generator.
The attack in both isn’t working right - rather than ‘fading in’ with a long attack, there’s just a period of silence (the length of which is controlled by the attack knob) and then the note plays as usual.
I’ve only got this connected up to a vactrol at the minute, as I haven’t built a vca yet. However, with a multimeter on the output, I get roughly a 4v signal when the notes playing. With longer release times, I can see this slowly trickle down. However, when I increase the attack it doesn’t slowly trickle up to 4v. Instead, nothing happens, then all of a sudden it immediately jumps to 4v. The length of time nothing happens for prior to jumping up to 4v varies based on where I have the attack set to.
If I connect my multimeter to the envelope out, it looks like a slow fade in/out.
What are the chances this is a consequence of my vactrol choice? I only had vtl5c9 and vtl5c6 to hand.
The LED in the VTL5C9 has a forward voltage of up to 2.5 V. This means if you’re just taking the envelope voltage and applying it to the vactrol, with a series resistor presumably, then the LED won’t light up until you’re putting something like a couple volts into it. After that it’ll brighten rapidly with higher voltage.
That’s one reason not to try to drive a vactrol directly with a control voltage. Another reason is control voltage sources generally aren’t designed to supply the kind of current a vactrol takes. Maybe they can but best practice is to treat CVs as reference voltages, not power sources.
The better way to do it is to use an op amp to supply a current to the vactrol that’s proportional to the control voltage. Then as soon as the CV is nonzero, the vactrol LED starts to light, and its brightness is at least roughly proportional to the CV.
Like this:
Here the current through the LED is just equal to the CV divided by R_FB.
(Depending on how bright you want the LED, it might draw more current than the op amp is happy supplying. You can use a slightly more elaborate design with a transistor in that case.)
Hmm, that’s a very good point! I hadn’t thought about the forward voltage, thanks.
It’s all quite cramped in this module, I might just leave these installed as an option and just build a proper VCA. I just wired them to some extra jacks at the bottom of the envelope generator. They work fine for short attack, and delaying the sound could be a useful option…
Hi - would the circuit on post 7 on this link using an LM324 work? I’m thinking of just using an LM358 to drive one or two LEDs, rather than the string included here.
Also, the envelope’s output also sits at around -2v when not triggered. If I put a 1M resistor to ground on the gate output, it sits at around 0v when not triggered and doesn’t seem to affect the gate range noticeably. Is there any reason to not add the 1M resistor?
Presumably. But for one LED or maybe two you probably wouldn’t need the transistor, the LM324 can source 20 mA min, 40 mA typical.
None I can think of. It’s often good practice to have a 100k pulldown resistor on module outputs anyway. (For one thing, if a module is unpowered, then its op amp outputs are probably floating and can cause trouble for a powered module they’re plugged into.) Though I rather rarely see it done. And a pulldown after a series diode is definitely called for in many cases. I think this is one. You might want to add a note suggesting it to Annotated Simple DIY Envelope Generator.
SO on homemade vactrols , what color led is best, if white , doent that need 3.5 v to energize brightness? Do People use red , green amber , blue . I was thinking of an ldr with a 5k to 2meg range . BTW love your designs and all around content and knowledge on synth matters and electronics .
I have a couple of cheap knock-offs in clip-together casing that have red LEDs on the inside. For a separate project a few years ago I built a small enclosure with some sockets and swapped colours in and out until I got the response I wanted, it made a big difference.
This stripboard layout does work, though I’m not sure how beneficial it is.
My knock-off vactrols in the plastic case still don’t respond as expected, and the attack still functions as a pre-delay. They have a little hole on the back so I can see the LED fading in, so I wonder if it’s something to do with the characteristics of the LDRs in these that’s responsible for that effect.
However, I picked up some Xvive VTL5C3s which do respond as expected, and I can get a much shorter decay time with these in addition to the attack working better.
I am thinking about using the above design in a 4 voice paraphonic build I’m starting. It doesn’t really produce very linear results, but I was looking at other designs and came across the SHOOSH design by Damien Clarke, which follows the vactrol with a gain stage with some notes on how to improve linearity.
I wondered if this might help me to get a better response? The audio input to the vactrol would be buffered from the VCO output, so I just added on the gain stage following the vactrol.
What do you think? And have I made any silly mistakes with the wiring…?
Also, would adding a current limiting resistor between the transistor and vactrol LED affect linearity of the response?
