How do I use a vactrol / optoisolator as voltage controlled resistor?

[armstrom]

I did a quick search and while there are a number of discussions about vactrols on here none seemed to answer my question so here goes...

I have a 0-5V variable voltage source that I would like to use in conjunction with an optoisolator in order to replace a pot in my circuit. After reading the data sheets for the vactrol brand my understanding is that they perform like "current controlled resistors". Meaning the resistance of the LDR varies in relation to the current through the LED side of the vactrol. So, to my non-EE mind that imples all I have to do is select a current limiting resistor that yields the maximum current at 5 V (basic ohms law stuff) and then any decrease in my voltage source will cause a change in resistance of the LDR as desired.

So, is it really this simple? Do I need some sort of more complicated transistor based LED controller in order to create this voltage controlled resistor scenario I'm after?

Also, is there another, easier method to create a voltage controlled resistor? I know JFETS can do the job, but I'm not quite sure how to set them up in such a manner and considering the wide fluctuations in electrical properties from one JFET to another (even of the same type) would the vactrol approach be easier? I'm pretty sure it will have fewer components.

Thanks,
Matt

[R.G.]

It's really that simple - if you are willing for the control range to be from 1.4V(or more, depending on the LED) to 5.0V, not 0V to 5V.

The problem is that the resistance of the photocell is a function of (notice I did not say "proportional to") the LED current, since it's a function of the light that hits it and the LED's output is a function of (notice again) the current through it, not the voltage across it.

LEDs, like all diodes, don't conduct much at all until you hit their minimum forward voltage, then they conduct heavily with any more voltage. To get an approximately linear current in the LED with a voltage with a 0V minimum, you need some kind of driver which forces the *current* in the LED to be controlled by an input voltage. This linearizes things.

One trick that works that's not thought of much is that you can put the LED in the feedback loop of an opamp, set up as an inverter. Feeding a 0-5V voltage to the input resistor causes a current of Vin/Rin to go into the input, and the opamp output then pulls an equal current through the feedback device to keep the minus input at Vbias. Notice that you need a bipolar power supply, or some kind of offset above zero to make this work. You can also do one of the literally hundreds of voltage-to-current converters to make a current proportional to a voltage.

But you get to choose fast,simple, or cheap - any two. 

[armstrom]

Thanks for the info! I'll probably just try the current limiting resistor method. My voltage source is a sensor that will have a normal voltage output of around 2.5V, in theory it can never fully reach 0V or 5V. When the sensor is "actuated" one way or the other the voltage will vary within that range. While the sensor will likely be able to reach a voltage below 1.4V a normal operator won't be able to keep it there very long so the high resistance of the photocell shouldn't be a problem. If it is I can always wire a resistor in parallel with the photoresistor side of the vactrol to control the maximum resistance of my "virtual pot". It will obviously change the voltage to resistance function of the entire circuit but that likely won't make a difference to the interactivity of the device.. I'll just have to play around with it.

Yes, I'm being vague deliberately

[dustin]

Also, is there another, easier method to create a voltage controlled resistor? I know JFETS can do the job, but I'm not quite sure how to set them up in such a manner and considering the wide fluctuations in electrical properties from one JFET to another (even of the same type) would the vactrol approach be easier? I'm pretty sure it will have fewer components.

has anyone ever tried this with an OTA?  I bought a few 13700's for some synth stuff I was going to try and was reading through this datasheet and noticed a section about setting it up as a voltage controlled resistor (page 11)  http://www.national.com/ds/LM/LM13700.pdf  I'm pretty new at this stuff so I can't really help but would love to get someones opinion on using an operational transconductance amplifier for this.  Also I have a project where I need a dual gang pot replaced with voltage control.... might this work?

Cheers,
Dustin

[R.G.]

Yes, I'm being vague deliberately

S'OK. There's plenty of info there to make a good guess.