Voltage Varying Resistor

[adept]

Hey guys! I was having a late night wait on the toilet when I got to thinking electronics as usual. Some crazy idea popped into my head and what it was is how to make a voltage varying resistor. I've done a few minutes of googling and all I get is using Fet's on their pre switch-on; however, that isn't very fun, now is it. The resistance goes from a meg-ohm to ohm's over just a few millivolts.
So, is there any insane way to either slow that slope down or some other circuit for the job. I've been playing with the idea of using a zillion passives to basically make a traditional Kelvin voltage divider with FET's for the taps and using a 10 bit ADC to get the job done. It seems pretty clever, but probably severely over complicated!

[IanB]

What is it that you want? You need to explain more clearly.

[adept]

What I want to do is to make a resistor circuit that varies based on the voltage applied to two terminals. This hypothetical resistor, with interesting applications in programmable opamp gain, precision voltage divider's, and the like, should vary predictably and in a controlled fashion. Using a FET isn't a good solution because the resistance changes from a very low to a very high value over a tiny voltage range. (http://www.instructables.com/id/Make-a-Voltage-Controlled-Resistor-and-Use-It/ )This guy on Instructables has the right intentions, but his execution, I must say, is even more insane than mine. But his intent is approximately aligned with mine. Is there any reasonable way to vary resistance with voltage?

edit:
And after a few minutes of spewing all over my schematic editor, this is a back of the envelope sketch of my basic idea for this resistor:

(you should be able to click on the photo to enlarge to a usable size.... hope this helps clarify a bit for you  )

[IanB]

Do you want a voltage dependent resistor or a voltage controlled resistor?

A VDR is a two terminal device with a resistance that varies according to the voltage across the two terminals.

A VCR is a three or more terminal device with a resistance that can be controlled by an applied voltage.

Assuming a VDR, do you want it to be linear? If so, what tolerance on the linearity? What about response and settling time?

I can't give you an answer, but if you can express your question more precisely your Google searches will turn up better results.

[amspire]

What I want to do is to make a resistor circuit that varies based on the voltage applied to two terminals. This hypothetical resistor, with interesting applications in programmable opamp gain, precision voltage divider's, and the like, should vary predictably and in a controlled fashion. Using a FET isn't a good solution because the resistance changes from a very low to a very high value over a tiny voltage range. (http://www.instructables.com/id/Make-a-Voltage-Controlled-Resistor-and-Use-It/ )This guy on Instructables has the right intentions, but his execution, I must say, is even more insane than mine. But his intent is approximately aligned with mine. Is there any reasonable way to vary resistance with voltage?

Yes there are ways, but the best solution depends on the exact problem you are trying to solve.

One way is with PWM (pulse width modulation). You have a high frequency oscillator with a voltage control for its mark-space ratio. You use this PWM output to switch a resistor. So if the resistor is 1K, and it is switched into the circuit 10% of the time, its average resistance will be 10K.

This solution will work in some circuits - not in others.

If you need a particular solution, you will have to give us full details of the circuit that you need. There is a good chance we will suggest a very different solution.

Richard.

[adept]

@ Ian, sorry, I didn't refresh! but yes, a VCR indeed is much closer to the device I'm thinking of. Are you implying they actually MAKE these things? Because you should link me up with one because I can't find anything using my google.... But that would be really neat!

@amspire
I think you aren't understanding me the same as Ian because PWM is a varying voltage, not a resistor. Yes, I know I can limit a current by switching a circuit on and off with PWM, but that isn't a real resistor. What I mean is like a thermister but instead of varying with temperature, varying with voltage.

oh, and i'll put this here again too!

click for bigger

[SeanB]

Most resistors are voltage dependant to some degree, and you can have them so dependant that they become a pretty close approximation to a constant current source, even over a range from 6-600V. They are commonly found in the voltstick to drive the 2 diodes to indicate polarity.

[IanB]

Would something like this help you?

http://www.maxim-ic.com/products/digi_pot/

[adept]

I'll call that good Ian. That is utterly fascinating. It isn't particularly precise, but that is pretty much what I had in mind. I applaud you for your patience! TYVM

[IanB]

http://www.edn.com/design/analog/4368893/Use-a-photoelectric-FET-optocoupler-as-a-linear-voltage-controlled-potentiometer

http://www.vishay.com/docs/70598/70598.pdf

http://www.maxim-ic.com/app-notes/index.mvp/id/4051

[amspire]

@ Ian, sorry, I didn't refresh! but yes, a VCR indeed is much closer to the device I'm thinking of. Are you implying they actually MAKE these things? Because you should link me up with one because I can't find anything using my google.... But that would be really neat!

@amspire
I think you aren't understanding me the same as Ian because PWM is a varying voltage, not a resistor. Yes, I know I can limit a current by switching a circuit on and off with PWM, but that isn't a real resistor. What I mean is like a thermister but instead of varying with temperature, varying with voltage.

oh, and i'll put this here again too!

click for bigger

No, if you switch a resistor on and off with varying mark-space ratios (PWM), you vary the average resistance. It is a PWM resistance rather then a PWM voltage. As I said, this works in some situations - not in others. There is no single device that works as you describe for all applications, so you need to come up with an application specific solution. What is your application?

Richard

[TerminalJack505]

They make something called a resistive opto-isolator  (aka Vactrol) that might be close to what you're looking for.  It's basically just an LED coupled to an LDR.

[Simon]

 Basically any non switching use of a transistor is some sort of variable resistance application but there is no one solution fits all. What do you think a power amplifier output stage is ?

[T4P]

How about using a FET?