Why are MOSFETs better than BJTs in application of Voltage Controlled Resistor?

Question

Why are MOSFETs better than BJTs in application of Voltage Controlled Resistor ? Thanks for your responses.

Answer 1

The following refers to NPN BJTs and N channel FETs

It's really hard to get good accurate representations of collector current versus collector-emitter voltage for different levels of base current. It's a bit easier for JFETs because, traditionally they claim to encompass the functionality of a voltage controlled resistor. Anyway, here are the best three I could find and they show a similar story: -

Top left is a BJT. Top right is a JFET. Bottom left is a MOSFET and bottom right I have drawn my own picture of the saturation/ohmic region. Basically, what I'm trying to say is that all three devices can be used as a voltage controlled resistor (VCR) because the base/gate control can vary the slope of the graph and the slope is basically conductance and that, in turn translates to resistance.

One thing that makes a JFET much more superior is that the drain current and voltage can be reversed (made negative) and the characteristic shape of the resistances are largely similar. They are not hi-fi quality the same but good enough for controlling volume in a cheap transistor radio or amplitude in a sine wave oscillator.

BJTs wobble a tad when the voltage is reversed because you are then relying on the transistor being symmetrical and having the same gain. MOSFETs usually (but not always) have a parasitic diode which makes them unsuitable for negative voltages when used as a VCR.

I'd go for JFETs every day.

As a footnote don't believe that the BJT characteristic looks like this: -

It completely fails to show the ohmic/saturation region properly, Rather, it seems to indicate that there is a one value of resistance irrespective of base current.

Answer 2

Why are MOSFETs better than BJTs in application of Voltage Controlled Resistor ? Thanks for your responses?

If such a device should be able to act as a voltage-controlled resistor it should have a voltage-current characteristic that is (at least) similar to an ohmic resistor. In particular it should exhibit a symmetric behaviour in the vicinity of the origin (positive/negative current for positive/negative voltages). More than that, we reqire that I=0 for V=0.

It is to be noted that both requirements cannot be fulfilled with a bipolar transistor (BJT). This is because collector and emitter cannot be exchanged without a drastic change in voltage-current characteristic. More than that, for Vce=0 the current Ic does not cross the origin for any finite control voltage Vbe. Hence, the BJT cannot be used as avoltage-controlled resistor.

However the situation is completely different for FET´s in particular for JFET´s. These devices have another working principle than BJT`s - they can be seen as voltage controlled resistors within the so-called "linear" region (ohmic region) for rather small VDC values (up to 1V). And the resistance between D and S can be varied (controlled) with the voltage VGS.

Answer 3

I have 2 images above. The first one is Ic vs Vce which i captured in a lab at university. The second I took off the internet of a FET because i didnt have one. In the FET one you can see that all the curves split off before they come flat. Giving a R = V/I for those curves will give different resistances. However, in my BJT one, you can see that before the curves become flat, they're all relatively close together, this means that R = V/I for all those curves will be slightly different, but more or less the same. Therefore, thats why MOSFETs can be used as a variable resistor, whereas BJTs can't really. I hope this helps