Servo Controlled Potentiometer using Rotary Position Sensor

Question

I need digitally controlled potentiometers but I want real pots and not a digital potentiometer chip because I want to integrate this with conventional relatively high voltage circuits like those you might find in a guitar amp.

My current idea is to mate a Rotary Position Sensor with a Servo:

The Murata SV01 Rotary Postion Sensor costs $0.95 USD and it's basically just a potentiometer with a D shaped hole instead of a shaft:

A cheap micro servo costs $3.50 USD:

I don't want a motorized pot because they're huge and expensive by comparison and I don't need a shaft. And the servo is easily controlled using a PWM output of a microcontroller.

So the question is, is this feasible? I'm not an engineer but I'm just concerned with coupling the two devices, accuracy, control, robustness, etc.

If you think it is feasible, can you recommend where I might obtain the necessary hardware to mate the knurled plastic shaft of the servo with the 4mm D shaped hole. I was thinking a mini-PCB could serve as a base with screw holes for the servo mounts and something to solder the Rotary Postion Sensor to.

Answer 1

Yes, that should work fine. The rotary position sensor you have pictured is made for servos like this; if you open the servo you will find such a pot used for position feedback.

One concern is crosstalk. The PWM signal is in the audio range, so any signal that bleeds through will be audible. This can be engineered around. I suggest separate power supplies for the position controls and the audio. In particular, do not allow the ground signal to be shared between the position and audio circuits.

Answer 2

The only problem I foresee is the servo not being able to turn the pot. If the servo is weak (and those tiny ones are quite weak) and if the pot is stiff (some of those little pots can be) then the servo may not be capable of fine control. It may be unable to quite reach its target position, or worse, it may not move at all. As usual, the only way to know is to try it. Get the pot first, and if it seems to rotate fairly freely, then you'll probably be fine.

Answer 3

To answer your question, yes, it is feasible.

However, I agree with @phil-frost that this is not the best way of doing things, if not the worst. You might have fallen into the trap I often find myself in, when you come up with an idea and then stubbornly struggle to overcome one problem after the other instead of rethinking the whole approach.

For example, "relatively high voltage circuits" is well, too relative. If it is indeed high then potentiometer like that will be arcing and will die quickly. If it is not that high then there are literally thousands of solid-state components that can do the job more precisely and reliably.

Like $5 AD5290 digital potentiometer rated to 30V or $2 MCP45HV51 rated to 36V.

Or simple $1.50 optocoupler rated from 40V, like PS2701-1 all the way to 400V MOC8204M. Is that voltage high enough? Not only optocoupler can provide better precision than your mechanical solution, it will also isolate your controller from target circuit, just like your servo would but without huge chance for switching noise on the output.