Vactrols and replacing variable resistors with control voltages

In a lot of electronic audio circuits the response is controlled by a variable resistor rather than a control voltage (CV). If you want to adapt these circuits to use in a modular configuration then you often feel it would be nice to replace or augment the variable resistor with a control voltage input. In some cases this can be quite easy – the variable resistor might just be set up as a voltage divider between two DC levels and effectively the output of the wiper is a control voltage. However it is often the case that the variable resistor is actually manipulating a waveform inside the circuit. For example, in a lot of filter designs the resonance level works on a variable resistance controlling the amount of positive feedback from the output to the input and this isn’t easily replaced with a control voltage.

I recently got this question from the web:

I want to add external CV [control voltage] control to an existing circuit. It’s to control the resonance of a filter where there’s already a pot to do this manually. My first idea was to use a vactrol and run the resistance from that in sequence with the pot with the pot then acting as an offset whenever there’s CV applied. However there’s a few things I’m struggling with.

1) It’s always only going to be a positive offset, even if you feed a bipolar signal to the CV, being that the vactrol can’t output negative resistance 🙂
2) What’s the strategy for managing current limits going to the vactrol? I’m familiar with *reducing* current via resistance but what if you don’t actually know what the current is going to be? How do you bring it within a usable range?
3) The big question: is this the best way to achieve the original aim? Any other suggestions?

This question nicely captures the classic problems in this situation – there isn’t an easy general purpose way of changing a control voltage in to a variable resistance. The simplest approach, as suggested here, is to use a vactrol. For those not familiar with the term a vactrol is a light pointing at a light dependent resistor in a sealed unit. By connecting the control voltage to the light you generate a variable resistance that depends on the control voltage while keeping the control voltage electrically isolated from the resistance.

Vactrols are easy to use but have lots of limitations – they don’t have a well defined relationship between the CV and the resistance and the range of resistance values achieved may not match what you want in your application. Generally they also have quite slow response. There isn’t much you can do about these limitations. Some kind of preprocessing of the CV might help set the range of resistances achieved to better meet your needs. You will also want to introduce the light dependent resistor in to the circuit with some kind of additional adjustment (perhaps the existing control) to set the control-point it is working around. For example, to get a bipolar response (Q1 above) you can add a DC offset to the CV so that the bipolar signal becomes an alternating positive signal and then set the adjusting resistor to position the range of the output to be that you are interested in. You can try putting the vactrol in series with the adjusting resistor instead of in parallel.

As far as I know most vactrols will go down close to 0 Ohms resistance when the controlling light is fully-on. This gives you the lower bound for the resistance, and hence upper bound for the in-circuit current. As just about all variable resistors also go to 0 Ohms you can pretty much add a vactrol in series or parallel with an existing variable resistor from the point of view of the minimum resistance (Q2). If you don’t want the vactrol to drop to 0 Ohms put a resistor in series with it, or manipulate the CV to limit how large the signal feeding the vactrol’s lamp gets.

As for Q3 – no, sorry there isn’t a general way of replacing resistance-driven circuits with CVs. There are various circuits that are called voltage-to-resistance converters (eg see the LM13700 datasheet) but in reality these have complicated limitations on how they can be used and don’t fully isolate the CV from the rest of the circuit. Unless you know a lot about how the circuit you are modifying works and fully understand the limitations it is hard to retrospectively introduce these in to an existing design that uses a mechanical variable resistor. If you *really* want a solution you could have a servo-motor turning a mechanical variable resistor – I can remember a few hi-fi buff friendly amps that used this approach.

Normally though the best you can do is to experiment with different ways of using a vactrol and learn to love their limitations. Any technical defects are called “character” and in classic audio equipment people pay good money for them!