A VCA of sorts

richardc64

At some point you'll want to hear the results of interconnecting all these digital circuits, and you'll probably want some dynamic control of the amplitude. The first circuit provides amplitutude control of digital waveforms without resorting to OpAmps, OTAs, or (ugh) cmos in linear mode.

Basically, as CV rises, the rectangular wave from the digital output to be heard "stands up" at the junction of the two diodes and passes through the 100K resistor and the coupling capacitor. This method was used in paia's Stringz-n-Thingz and a similiar technique was used in their Cosmic EGG.

Two simple EGs are shown, which don't really need much explanation. Select pot and cap values for the desired durations. The Gate Inputs should probably be buffered by an AND gate with both inputs tied together. If the output is to be routed many places, it can be buffered by a single-supply OpAmp such as the LM324.

vca-env.gif

Description:

Filesize:

2.88 KB

Viewed:

63745 Time(s)

widdly · Posted: Wed Feb 13, 2008 6:55 pm Post subject:

Rather than an opamp buffer, would it be possible to use a 4050 cmos buffer?

I think this could be implemented in a general sort of a fashion so it could be used whenever you want to route one thing to a lot a places.

Tim Servo · Posted: Wed Feb 13, 2008 9:19 pm Post subject: A VCA of sorts

I've thought of a slightly different approach to a "Logic VCA" (LVCA ©)...
Wink

You apply the digital signal to the control pin of an analog switch, and apply the CV to the INPUT of the switch. Now, at the output of the switch you're getting whatever level the CV is at, but turned on an off by the digital signal. The effect is that you get a signal with the frequency of the digital signal, but the level of the CV. The nice thing here is that you can use a bipolar CV and get a bipolar output signal (just don't apply a bipolar signal to other logic; use a diode to chop the negative portion off). I've designed something like this into my version of Ian Fritz's double pulser, but I haven't breadboarded it yet. Might also be fun with a multiplier and/or divider ciruit to get CV control of output aplitudes (and polarity). Like I say, I haven't tried this yet, but I'm pretty sure it would work.

Tim (might be fun) Servo

richardc64 · Posted: Thu Feb 14, 2008 6:26 am Post subject: Re: A VCA of sorts

Pretty cool.

Pehr · Posted: Thu Feb 14, 2008 7:55 am Post subject: Re: A VCA of sorts

Tim Servo · Posted: Thu Feb 14, 2008 12:24 pm Post subject: A VCA of sorts

Ah, I'd forgotten about the switching level requirements for CMOS analog switches. You're right: on a 4016 / 4066, the "logic" for turning the switch on and off has to swing between the supply rails (or close to it). So if you're powering an analog switch with ±7.5V so that you can switch a ±5V analog signal, the switching input has to flop back and forth between +7.5V and -7.5V. Still, might be worth looking for a simple level shifter for the switching input.

The 74HC4316 looks like an ideal solution. This part is still in production, and can be had for under a buck. Also looks like it would be happiest running on ±6V supplies (still allows switching a ±5V analog signal with 'standard' 0 to 5V logic). I think this would be something worth looking into, as I think I'd want my Lunettas to be able to play nice with my other modular stuff.

The TL60x series is interesting, but they only have one or two switches per package (with only a single control pin), AND it doesn't look like they're cheap. Might not be in production anymore.

Tim (looks interesting, but definitely NOT cheap) Servo

mosc · Posted: Fri Feb 15, 2008 9:23 am Post subject:

I think this discussion is great but I caution people who are just starting out. These kinds of circuits will require a lot of experimentation and depending on the electrical characteristics of the input signals will work differently.

For beginners, just try to work with the OSCs, counters, shift registers, gates and other digital circuits that don't require too many resistors, diodes, or capacitors to work.

I don't want to see people get the feeling that you have to be a circuit designer to get these Lunettas to work and to have fun.

That said, please continue.

Scott Stites · Posted: Tue Feb 19, 2008 7:26 am Post subject:

All I can say is whhhooooaaaaaaaaaa.....how did I miss this entire forum popping up? I need to wake up....thanks for this, Mosc, what an excellent idea...

Anyway:

Scott Stites · Posted: Tue Feb 19, 2008 7:31 am Post subject:

Badda-Bim, Badda-Bam:

http://www.cgs.synth.net/modules/cgs28_seq_switch.html

Scott Stites · Posted: Tue Feb 19, 2008 7:41 am Post subject:

Hmmm....I guess he's still using -0.6V and 14.4V for the rails, but the supplies aren't too complicated.

I've configured bipolar CD4066s using a dual op amp for the supply. In that case, IIRC, I used two 10K resisors to divide the +15V rail to 7.5V, then fed that to the dual op amp - one section was a voltage follower and the other was an inverting buffer. The outputs drove the CD4066s just fine.

This arrangement of Ken's looks a bit better - you have a wider range to play with, so you don't have to worry quite so much about exceeding the input level.

Cheers,
Scott

Pehr · Posted: Tue Feb 19, 2008 7:42 am Post subject:

richardc64 · Posted: Tue Feb 19, 2008 8:38 am Post subject:

Scott Stites · Posted: Wed Feb 20, 2008 2:05 pm Post subject:

mosc · Posted: Thu Feb 21, 2008 1:13 pm Post subject:

Hey, you guys are getting too complicated, or at least this talk of full-range bipolar switches should not be in this forum.

Using the CMOS decoder/switches is great, just use them to switch digital sigs.

fluxmonkey · Joined: Jun 24, 2005 Posts: 708 Location: cleve

mosc · Posted: Thu Feb 21, 2008 4:43 pm Post subject:

Good point, Bbob. Historical accuracy is not important at all, IMHO.

As for schematics, there really were no schematics. We just wired chips up direct to the banana plugs - for the most part.

23isgood · Posted: Thu Feb 21, 2008 6:03 pm Post subject:

Really? Thats it? No resistors or diode protection? Im adding some protection circuitry to mine. I want to be absolutely sure that whatever I plug up to these circuits will not blow them up. I do plan on interfacing mine with my modular gear, for sequencing. So on a proto board you could have a few cmos chips on there. I guess the majority of the work is in the case.

mosc · Posted: Sat Feb 23, 2008 12:09 pm Post subject:

Well, CMOS chips these days are much better than in the past for ESD protection. Still, it's not bad to be careful.

jnuaury · Posted: Tue May 13, 2008 6:17 pm Post subject:

im think slow vactrols are a nice match for the lunettas

this of course only gives you percussive envelopes but i think that is actually enough for the lunetta style

otherunicorn · Joined: May 11, 2008 Posts: 136 Location: Australia

stolenfat · Posted: Sun Dec 28, 2008 2:45 am Post subject:

maybe i missed it while reading over this forum, but using the schematics at the top to send to a VCA... is there a simple way to do that? eg an easy circuit in the lunneta style that would take the digital audio signal and amplitude modulate it in accordance with the AR signal?

richardc64 · Posted: Mon Dec 29, 2008 6:22 am Post subject:

The circuit at the top IS a VCA. Can't get much simpler than a couple of diodes and resistors or, as Mr. Servo suggested, 1/4 of a 4066.

stolenfat · Posted: Mon Dec 29, 2008 2:05 pm Post subject:

oh i see! that is incredibly simple.. wow

one question, theres a cap in the top schematic, does it have to be a specific value (i'd gues 10uf) or can you tinker with it- it seems so simple yet so delicate.

blue hell · Posted: Mon Dec 29, 2008 2:13 pm Post subject:

adammokan · Joined: Apr 19, 2009 Posts: 2 Location: Chandler, Arizona

Hey everyone. I posted this question over at the Deathlehem forum, but wanted to throw it out here, as well...

I'm trying to run a simple 4093 output into the dual-diode VCA. Signal doesn't seem to be hot enough when I check the anode side on that first diode. So I'm running it through a 4049 to attempt boosting the signal. Still extremely faint, but I may have something wrong with my resistors on the feedback loop of the 4049. I forget the math there. Positive my oscillator is working because it sounds fine on the cathode side.

Either way, I need to take a break for a few hours and just wanted to make sure I'm heading in the correct direction and that my theory of simply running the output of the 4093 is not hot enough to be audible through this VCA without pre-amping... or should I just throw an amp on the output of the VCA? 741, etc?

Trying to keep it simple.

Thanks!!!