4
\$\begingroup\$

I am trying to make a variable resistor using a MOSFET, but the problem is that I don't want it to vary from voltage, but instead current. I understand that MOSFETs work by creating an electric field from the gate. I don't understand why the gate is not influenced by current.

Aren't electric fields influenced by current? If so, why aren't MOSFET gates increased with current?

All in all, my question is: why do MOSFET gates only change when you change voltage, and not current?

\$\endgroup\$
3
  • 1
    \$\begingroup\$ Please can you edit your question to explain what problem you're trying to solve or what circuit design you have, with any schematic. This could be fundamentally an XY question. Thanks. \$\endgroup\$
    – TonyM
    Commented Sep 24, 2022 at 7:39
  • 5
    \$\begingroup\$ ”Aren't electric fields influenced by current? No, by voltage. Magnetic fields are related to current. \$\endgroup\$
    – winny
    Commented Sep 24, 2022 at 9:37
  • \$\begingroup\$ Basically no different from a triode vacuum tube. \$\endgroup\$
    – Hot Licks
    Commented Sep 25, 2022 at 2:50

3 Answers 3

6
\$\begingroup\$

I suppose you could just tie the gate to ground with a resistor and inject a current. The current will be turned into a voltage that will appear across the gate. A resistor is a current to voltage converter after all.

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
6
  • \$\begingroup\$ If I input to "in" 1 amp, what voltage would it become (in relation to R1?) \$\endgroup\$
    – gbe
    Commented Sep 24, 2022 at 2:59
  • 3
    \$\begingroup\$ 100 V. It will also consume 100 watts. Why do you want to use such a large current as a control signal? \$\endgroup\$
    – The Photon
    Commented Sep 24, 2022 at 3:11
  • \$\begingroup\$ @The Photon No it was just an example because 1 amp is a very simple unit. Would I multiply 1 amp by 100 ohms to get the voltage? Is that the equation? \$\endgroup\$
    – gbe
    Commented Sep 24, 2022 at 3:18
  • 2
    \$\begingroup\$ @Walter You would need to size your resistor so your operating current range matches that of the linear region of the MOSFET. If you produce too little gate-source volage the MOSFET will not conduct and if you have too much the MOSFET will just behave as a near short-circuit. A the MOSFET behaviour is not linear either. Don't expect something easily predictable. \$\endgroup\$
    – DKNguyen
    Commented Sep 24, 2022 at 3:30
  • 5
    \$\begingroup\$ I was always talking about MOSFETs. BJTs respond to base-emitter currents directly. But BJTs don't behave as resistors really. \$\endgroup\$
    – DKNguyen
    Commented Sep 24, 2022 at 3:32
9
\$\begingroup\$

A MOSFET gate is essentially a capacitor that creates a field which allows current to flow through the body source and drain, and as the voltage increases, the body resistance decreases. The principle is similar to a vacuum tube, where grid voltage allows electrons to flow from the cathode to the positively charged plate.

Some current flows due to charging and discharging the gate capacitance, and some very small leakage current, but conductance is a function of gate voltage with respect to the source.

You can add a resistor from gate to source, and the current will provide a gate voltage and this will vary the resistance between drain and source, but it more a voltage controlled conductance.

MOSFETs can operate in a linear triode or ohmic mode, or in saturation or active mode:

https://en.wikipedia.org/wiki/MOSFET

\$\endgroup\$
6
  • \$\begingroup\$ Is there any way I can do this that responds to current? Like I don't want the gate to respond to voltage. But instead if I increase current, the resistance decreases \$\endgroup\$
    – gbe
    Commented Sep 24, 2022 at 2:11
  • 1
    \$\begingroup\$ @Walter if you want it to respond to current, you want a BJT. \$\endgroup\$
    – Hearth
    Commented Sep 24, 2022 at 2:17
  • \$\begingroup\$ @Hearth Yes, but a bjt as I understand it, only limits the output. The BJT has a set gain. If gain is 100, and the base current is 1 amp, the max amount that could flow out the emitter is about 100 amps. Am I understanding it wrong? \$\endgroup\$
    – gbe
    Commented Sep 24, 2022 at 2:23
  • 1
    \$\begingroup\$ @Walter I suppose you could just tie the gate to ground with a resistor and inject a current. The current will be turned into a voltage that will appear across the gate. A resistor is a current to voltage converter after all. \$\endgroup\$
    – DKNguyen
    Commented Sep 24, 2022 at 2:47
  • 2
    \$\begingroup\$ @Walter A MOSFET has a set gain as well; it'll be listed in the datasheet as \$g_m\$ or \$g_{fs}\$. If \$g_m\$ is 10 S, then a gate-source voltage of 1 V above the threshold voltage means you'll get 10 A of output current. \$\endgroup\$
    – Hearth
    Commented Sep 24, 2022 at 5:35
3
\$\begingroup\$

First off I want to applaud you for asking a very interesting question, you're trying to connect the EM physics to the circuit theory and that's not easy and it's good youre thinking like this.

Electric fields are influenced by charge, not current. You might say current is moving charge, but in the DC case you mean it's a constant stationary current. Take a look at Maxwell's equations. It doesn't use current, it uses current density vector J, which is about charge.

Here's an analogy. Applying voltage is like taking a heavy box and lifting it onto a table. Current is like pushing the box across the floor. The MOSFET only cares about how many boxes it can see on the counter.

\$\endgroup\$
5
  • 3
    \$\begingroup\$ How is J any more about charge than I? \$\endgroup\$
    – Hearth
    Commented Sep 24, 2022 at 5:32
  • 2
    \$\begingroup\$ I is a 1D scalar value, J is a density vector in 3D space. They're fundamentally very different. \$\endgroup\$
    – Shredder
    Commented Sep 24, 2022 at 7:44
  • 1
    \$\begingroup\$ I'm not contesting that they're different. I'm asking how J is more related to charge than I, which seems to be what you're saying in the second paragraph. \$\endgroup\$
    – Hearth
    Commented Sep 24, 2022 at 13:29
  • \$\begingroup\$ Current I is agnostic of the size of the wire/channel and net velocity of the charges. Given a certain current I, that could be a small amount of charge passing through very quickly, or a huge amount of charge passing through slowly. Same current, different charge, and therefore the second will generate a stronger electric field. J keeps this info. \$\endgroup\$
    – Shredder
    Commented Sep 24, 2022 at 23:46
  • \$\begingroup\$ A current doesn't generate an electric field, it generates a magnetic field. \$\endgroup\$
    – Hearth
    Commented Sep 25, 2022 at 0:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.