# MOSFET as a voltage controlled resistor

This question might be too localized, but I try.

Is it possible to replace a variable resistor by a MOSFET, under conditions shown in the following schematic?

If yes, can someone propose a MOSFET type or the required MOSFET parameters.

simulate this circuit – Schematic created using CircuitLab

Update

What I am actually trying to accomplish is to replace R2a by something simple that I can control with a microcontroller (DAC).

I am hacking an existing device and can not replace the resistor R1.

• You seem to be presupposing a solution. Is the output supposed to be the specific voltage range (0.12 to 4.00 V) or is it supposed to be proportional to the (nominally 12 V) supply voltage? In other words, does it really need to be implemented as a voltage divider as shown? – Dave Tweed Jun 10 '13 at 13:49
• The output is supposed to be proportional to the supply voltage (10V .. 14 V). The resistor R1 is already implemented I can not change that. I need to replace R2 by somethig simple, that I can control by a microcontroller. – sergej Jun 10 '13 at 14:14
• Pop up two levels and explain what you are really trying to accomplish. Stick to what you want to accomplish, not how you think it could be accomplished. By presenting some imagined solution, it makes getting to the real problem harder, especially since we don't even know what it is. – Olin Lathrop Jun 10 '13 at 14:14
• have a look at graffiti.virgin.net/ljmayes.mal/comp/vcr.htm – JIm Dearden Jun 10 '13 at 14:19
• If you're hacking an existing device, get a 1 kOhm 4-way digital potentiometer, and gang it up to get you a 0 - 250 Ohm digitally controllable resistor. You control it with I2C or SPI, digitally, rather than with a DAC. AD5254 or AD8403 could probably do it. – Jon Watte Jun 10 '13 at 17:42

Technically the MOSFET can operate as a variable resistor, but there are two main issues:

1. In the ohmic region (which is quite narrow, in terms of output voltage) the linearity is poor, and it also depends on input voltage. It won't be very easy to tune it to behave like a proper resistor.

2. MOSFETs' output resistance is usually not an accurate value, and it will be hard to get the exact value from the datasheet. What you can do is to measure it for various input and output voltages, and to create a table with the values. But if you don't need it to be accurate, you can use the graphs in the datasheet.

Another choice can be to use an integrated VCR.

• +1 for actually answering the question. Not that all the other comments are not valid, of course. – Bobbi Bennett Jun 10 '13 at 16:23
• According to Martin Hartley Jones' venerable textbook p. 139, linearity issues can be overcome by using negative feedback. Furthermore, the same textbook says that FET voltage-controlled atenuators were commonly used as automatic gain control devices (dynamic range compression) for example by Dolby. Siliconix' note AN105 has some example circuits with JFETs voltage control resistors: a feedback circuit and voltage-control of an operational amplifier's gain, but not the fabled Dolby circuit. – Fizz Oct 6 '14 at 16:00
• Also, another name for the voltage-controlled resistor is transconductance amplifier, which is more likely to be used to refer to an operational (rather than single FET) amplifier. – Fizz Oct 7 '14 at 1:37
• Three circuits implementing a voltage-controlled resistor using an OTA (operational transconductance amplifier) are given in the data sheet of LM137000, pp. 12-14. – Fizz Oct 9 '14 at 19:00

Yes Sergej. You are 100% right !

A MOSFET can easily be used as a variable resistor. You have to consider few important parameters before using as a variable resistor. Main things are

1. The minimum resistance you need and the $R_{DS(on)}$ of the MOSFET you have chosen.

2. The MOSFET's behavior in the linear region though it is similar for almost all the MOSFETs.

Now we will see how we can use it as a variable resistor by making use of following characteristic curves of MOSFET

1. When the $V_{GS}$ is below $V_{th}$ of the MOSFET it is in a cut-off mode that means the entire supply voltage comes appears across MOSFET.It means that now the MOSFET is acting as a open-load with infinite resistance.

2. When you slowly increase the gate voltage the MOSFET slowly starts conducting by entering the linear region where it starts developing voltage across it which we call as $V_{DS}$ . In this region the MOSFET acts as a resistance of finite value.

3. Now when the MOSFET enters the saturation region the resistance of the MOSFET is the least and is equal to the $R_{DS(on)}$ of the MOSFET which is mentioned in the circuit.

• I think it would be more accurate to say in the linear region the MOSFET "acts something like a resistor". If it were really like a resistor, those voltage-current lines would be straight lines, but you can plainly see they are not. They approximate straight lines, over some small area, but to say they are like a resistor is an over-idealification. – Phil Frost Jun 10 '13 at 17:21
• Yes Phil.I completely agree with you.In my answer i just wanted to give an idea as in how its range can be viewed like a resistance. – Durgaprasad Jun 10 '13 at 17:42