Electronics > Beginners

Is there such a thing as a variable resistor / transistor combo?

(1/3) > >>

Plasmateur:
I'm looking to vary the resistance in a 10 Watt circuit.

Everything on Mouser in the 10 Watt range is called a rheostat and is around \$50+ to buy.

Can't I just use a low wattage potentiometer and a transistor to achieve nearly the same effect of a higher wattage variable resistor... minus some linearity?

Is there such a device that is sold as an all-in-one package, or is this usually a build it yourself kind of thing?

EDIT: FOUND THE TERM: Adjustable Voltage Regulator

Whales:
Yes and no.

I use the exact circuit you describe to power the LED strips above my desks.  12V in and an N-channel mosfet controlled by a pot.  Mosfet is wired in common-source arrangement (12V -> LEDs -> Mosfet -> GND).   The non-linearity is actually really useful, because it approximately counter-acts the non-linearity of the LEDs and human eyes.  I make more of the pot's rotation range "useful" by adding some resistors either side of it.

List of things to be wary of when using transistors as variable resistors:

* Many transistors are not rated for DC resistive use, they're rated for switching (fully off and fully on) instead.  Check the graph on the datasheet called "safe operating area", make sure there is a DC rating on it and see what the max conditions are.  Prepare to be disappointed for most parts.

* Common source N-mosfet circuits are very nonlinear (input V versus impedence or output V)
* Common emitter NPN BJT circuits are nonlinear too, but possibly less so?  See their graph of beta/hFE/currentgain on their datasheet (input I versus output I).

* You might be able to get away with a source-follower (emitter follower) circuit in many circumstances instead.  It's approximately an adjustable voltage source made out of a transistor and a pot.  Linear relationship between input voltage and output voltage, with a bit of load-dependent error.  Really useful to know, especially for controlling small brushed DC motors or lights with in-built resistors.

* Some jfets can be used very linearly as variable resistors.  Alas Jfets tend to be too small for power usage.

* As transistors heat up they change gain (thus changing their simulated resistance).  In the worst cases this can cause them to perform thermal runaway; but you can make choices to avoid this happening.

Plasmateur:

--- Quote from: Whales on December 01, 2021, 09:36:07 am ---Yes and no.

I use the exact circuit you describe to power the LED strips above my desks.  12V in and an N-channel mosfet controlled by a pot.  Mosfet is wired in common-source arrangement (12V -> LEDs -> Mosfet -> GND).   The non-linearity is actually really useful, because it approximately counter-acts the non-linearity of the LEDs and human eyes.  I make more of the pot's rotation range "useful" by adding some resistors either side of it.

List of things to be wary of when using transistors as variable resistors:

* Many transistors are not rated for DC resistive use, they're rated for switching (fully off and fully on) instead.  Check the graph on the datasheet called "safe operating area", make sure there is a DC rating on it and see what the max conditions are.  Prepare to be disappointed for most parts.

* Common source N-mosfet circuits are very nonlinear (input V versus impedence or output V)
* Common emitter NPN BJT circuits are nonlinear too, but possibly less so?  See their graph of beta/hFE/currentgain on their datasheet (input I versus output I).

* You might be able to get away with a source-follower (emitter follower) circuit in many circumstances instead.  It's approximately an adjustable voltage source made out of a transistor and a pot.  Linear relationship between input voltage and output voltage, with a bit of load-dependent error.  Really useful to know, especially for controlling small brushed DC motors or lights with in-built resistors.

* Some jfets can be used very linearly as variable resistors.  Alas Jfets tend to be too small for power usage.

* As transistors heat up they change gain (thus changing their simulated resistance).  In the worst cases this can cause them to perform thermal runaway; but you can make choices to avoid this happening.

--- End quote ---

Thank you for all of this information. Luckly, I'm right at 12V, so if this is in your ballpark for something you've already done... hopefully I'll find a few options in the safe operating range that will work for me.

ledtester:

--- Quote from: Plasmateur on December 01, 2021, 09:26:27 am ---...
Can't I just use a low wattage potentiometer and a transistor to achieve nearly the same effect of a higher wattage variable resistor... minus some linearity?
...

--- End quote ---

What's your application? Sounds like you are looking for a current limiter or constant current circuit -- like with an op-amp, low-wattage pot and a transistor (typically a MOSFET but a BJT could also work) which dissipates the power, e.g.:

https://www.bristolwatch.com/ele3/2a.htm

ambrosia heart:
I'm looking to vary the resistance in a 10 Watt circuit.

Seek help from taobao  ;) and ask Americans living in Hong
Kong or TaiWan to buy and sent to you  :P . It is 25W.  :popcorn: See photos.