I’m trying to make a VCA with as few components as possible, normally the CV part requires an additional op-amp.
I’ve come up with a compromise that’s not too bad: I use a resistor (4.7k on the OTA’s Ibias input) as a voltage/current converter.
A voltage divider pot between -12v and +12v if nothing is plugged into the CV input to control the gain.
If a jack is plugged into the CV input, then the voltage divider is connected between -12v and CV (which varies between ±12v) and acts as an attenuator for the CV.
The thing is, the falstad simulation works like a charm and I’m within spec according to my OTA datasheet, but it seems a bit too good to be true.
Does anyone see anything wrong with this set-up?
I just got out of bed and so probably shouldn’t comment but… it looks to me like you’d have gain not proportional to CV. The control current pin of the LM13700 is I think one or two diode drops above VEE (-12 V), so when CV (after the attenuator) is 0 V there is a voltage drop across the 4.7k, a control current flows, the gain is nonzero.
Here is a VCA from the Barton Analog Drum module:
The 10k after the voltage follower has its downstream end a diode drop above ground, so to a first approximation 0 V CV gives no current and the VCA gain is zero. Really the gain doesn’t become nonzero until the CV is above a diode drop, so it’s still not perfectly proportional, but it’s fairly close.
Damn yep you are right, at 0V i get a gain of 0,78 i didn’t think about that.
Thank you =)
Well the simulation show me that i can get a zero gain at 0v by tweaking the 33k for a 26k resistor at the audio input but it still seem odd.
I guess it’s breadboard time !
You can build a VCA using both OTAs in the LM13700 without any op amps as in David Haillant’s Simple VCA. That’s probably as simple as it can get using an OTA IC and nothing more.
(a) Simply a fun challenge
(b) To save space
(c) To save money
If it’s (a), have at it on the breadboard, and good luck!
If it’s (b) and/or (c), however, I have some suggestions. As you hint on your sim, you kind of need to buffer the CV input. You could do this with discrete components, of course, but you’ll get better results with an op-amp, so the idea that the op-amp is additional becomes a bit shaky.
The LM13700 is a dual OTA, and as the truism goes, you can’t have too many VCAs, so building a dual VCA satisfies (b) and (c). You can do it with a TL074, a TL072, and a handful of other components. I’ve built it comfortably in 3U x 5HP. My design is based on Thomas Henry’s VCA-1. VCA-1 (birthofasynth.com)
Here’s my schematic:
I took @plop at their word when they said “a [singular] [OTA*] VCA with as few components as possible”, and I think the circuit pointed out by @K.ostas probably comes close. It uses the other OTA on the chip to buffer the CV, and one of the Darlingtons on the chip to buffer the output, and the signal input doesn’t really require buffering. So a single chip and a few resistors does the job. The Thomas Henry dual VCA, I would guess, probably does the job better, but it’s three chips, two transistors, two diodes, four capacitors, and some resistors to make two VCAs — definitely not as few components as possible.
* Of course you can make an even simpler VCA without an OTA — a single transistor will do — but the OTA was specified, and it’s likely to work much better than the single transistor.
Yep, so my goal is to use an dual-OTA for making a dual VCA, and i’m mostly tweaking the schem’ for fun and come up with somthing new (but still functional) i know there is plenty of OTA based VCA in the wild and i can just copy a schematic but that’s not my goal here.
I want it to use less component as possible to use it as an input and/or output stage of effect module mostly to save PCB’s space, i will try different VCA desing to get the better space/usability ratio.
My understanding is that you can easily make a dual VCA using a dual OTA but you will have to add a dual op amp at least to buffer the CVs inputs. If you must use one and only one IC (but why?) then, the Haillant design is the most simple OTA VCA that I know of. Either way it’s as KISS as it gets.
And between an OTA and a simple transistor, if one has the time and patience to experiment, there are probably at least ten other designs based on almost an equal number of different ICs that sound much better than the transistor VCA and probably as good as an OTA VCA.
i didn’t think about that.
