# Is there such a thing as a resistance oscillator?

I think I understand how a Oscillator works (VCO). One uses an Oscillator when one wants for example a square wave oscillating between sy, $0$ volts and $5$ volts.

My question is: Is there such a thing as a resistance oscillator?

That is, how can one create a circuit where it is the resistance that changes as a function of a time? Can this be done as a square wave? Sine Wave?

• You mean like a programmable electronic load? – Wesley Lee Aug 12 '18 at 1:04
• @WesleyLee: Yes, I think that is what I mean. Something that will change in resistance over time. – Thomas Aug 12 '18 at 1:30
• Well.. You can always convert a resistance to a voltage and then feed that voltage into a VCO. – Harry Svensson Aug 12 '18 at 1:51
• A voltage controlled resistor (VCR) has been recently discussed but if you want to control voltage controlled resistance with a bias voltage you can use a current source with sense feedback.(V/I(t)=R(t)) But people usually just use a fixed V or R and modulate current. – Tony Stewart Sunnyskyguy EE75 Aug 12 '18 at 2:10
• Through the wonders of Ohm's Law, we can convert a resistance to a voltage. VCOs are often controlled by a resistor or a potentiometer. – Lundin Aug 14 '18 at 9:16

A VCO is an oscillator that uses a variable DC voltage to change its frequency. The design of the circuit and its power supply voltage determine its high output voltage and its low output voltage. If you want to change its output level then you need a voltage controlled amplifier (VCA).

Do you know that a variable resistor is commonly used as a volume Control? It is a variable voltage divider. But instead of audio as the input you could use a DC voltage that is changed when you turn the variable resistor. Connect +5V to one end and 0V to the other end. The center pin is the output which goes from 0V to +5V and it can control the frequency of a VCO.

It is easy to change a voltage as a function of time with a resistor connected to a fixed voltage that charges a capacitor. The voltage rising in the capacitor can control the resistance of a transistor or a Mosfet.

There are several different types of voltage controlled resistor.

• One is the "Vactrol" whish is a photoconductive element illuminated with a controllable light-source - often an LED shining on a light-dependant resistor.
• Another is the "Carbon Button Amplifier" which uses a voice-coil to vary the pressure on a pressure-sensitive carbon contact (which varied the resistance of the contact). Long ago these were used to amplify telephone signals on long-distance wires.
• Another is to use a field effect transistor. FETs behave somewhat like resistors at small signal levels.

These methods are all becoming less common, and electrically controllable variables resistors are simulated by switching fixed resistors in and out of circuit - this technique is used in digital potentiometers etc.

If you take a time-varying signal from an oscillator and feed it to a vactrol you'll get a time-varying resistance.

One can gradually collect voltage to a capacitor with a current or electric current to an inductor with a voltage. Having one quantity A which grows or decreases gradually due another quantity B is essential for oscillations. In addition we need a control circuit which reverses the polarity of B to force A in turns to increase and decrease. The control circuit must directly read A to decide how B should be changed.

We have several mechanisms where resistance increases or decreases due some quantity X. For example a voltage over a resistor makes a resistor warmer due the dissipation. The resistance of an adjacent NTC resistor decreases gradually. As well a voltage can run a DC motor which turns a potentiometer.

These can cause oscillations if we allow some test signal to be applied to measure the resistance. Resistance cannot be seen directly, a voltage is needed to see what current is caused or a known test current must be made and the voltage is measured.

A simple resistance oscillator with an electric sensing circuit can be a bridge, where a motor turns a pot and that pot supplies a voltage to motor from a bipolar DC source. Without any inertia the motor stops to the position where it gets 0V, but the rotating mass easily continues the movement and the balance is never found.

If we need a system where the resistance swings and at the same time just that resistance must not be used for any electric signal in the oscillator, but it still must be the seen quantity of the circuit which controls the resistance, we unfortunately are out of the luck.

normally you should always have some imaginary part for an oscillation. Something ideal, that is only resistive will never oscillate.

Maybe this can help. So it is necessary to have a resonance for an oscillation. This is only possible with introducing an inductor and a capacitor as well.

A 1mm cube of silicon has a thermal time constant of ...... 11.4 milliSeconds.

You could thermally-bond a resistor to be heated, and a thermistor that senses the heat and moves the output of a voltage-divider above or below some threshold.

Why use silicon? Some resistors have cores of clay, or ceramic, and thus will propagate heat similar to silicon.