Hi Sayeed Ahmad,

I think we can't classify like that, because a BJT can be used as VCVS,CCCS,VCCS,CCVS amplifiers. We can say that MOSFET is voltage controlled device because its drain current is dependent on gate-source voltage, and similarly a BJT is a current controlled device because its output is controlled by input current.

The big difference between the two is the way they are controlled:

*** MOSFET: When you apply a voltage to the input, you are building up GATE charge (for example, 11nC for IRFZ14). So, although you are controlling it with VOLTAGE, it is more accurate to think of it as CHARGE controlled. In other words, you can't instantly increase the gate voltage, you need a certain amount of charge (11 nano Coulombs in the IRFZ14 example to fully turn it on). During when the gate charge is increasing, at some point, you will hit Vgs=Vth, at which point the MOSFET will start conducting. Up to that point, you have built up a certain amount of charge. Typically, you do not stop at Vth, since, the Rds (ON resistance) is pretty lousy at that point. You want to continue and increase the Vgs more, so, you turn on your MOSFET harder. However, since the MOSFET starts turning on right at Vth, you hit a MILLER PLATEAU. In other words, Above Vth, you have to start pumping a lot more charge into the gate to increase the Vgs (why, since you are charging the Cgd as well as Cgs at that point). Above the Miller Plateau, it becomes easy to build up Vgs again, since Cgd is now charged up. Above 10V or so, it doesn't make a big difference how much more charge you put into the gate, Rds will change minimally.

*** BJT: Since BJT's Ic goes up exponentially with the increasing Vbe, what makes the difference above, say, 0.7V or so, is the base CURRENT. So, although it is the voltage that is turning it on and off the BJT, you need to increase the Ib incredibly to go above Vbe>0.7V. But, since Ic=hFE*Ib, the amount of Ib you pump into the base start making a big impact on Ic, and, becomes the dominant control element. Vbe changes so minimally at this point that, it can be considered as an irrelevant parameter. So, in that sense, BJT's can be thought of as CURRENT CONTROLED devices, though, this sentence needs careful thought before it is tossed around as a learning material.

One more big difference between MOSFET and BJT:

When MOSFET is ON, it is ON forever ... When a BJT is ON, you have to keep the same Ib current flowing to keep it ON. Otherwise, it will turn off.

Hope this helped.

Hello Tolga,

Good to see you back again!

I appreciate your answer particularly the first section - MOSFET.

But I am really concerned about your last paragraph -

1) "When MOSFET is ON, it is ON forever...."

------> What if we withdraw the gate voltage Vgs?

------> Will the MOSFET remain ON forever?

------> Then can we apply only a single pulse to turn it ON?

------> How will we turn OFF the device?

2) "When a BJT is ON, you have to keep the same Ib current flowing to keep it ON. Otherwise, it will turn off."

-------> Suppose my Ib is 60uA initially, increased to 75uA and then reduced to 40uA. Will my BJT remain ON or it will get turned OFF?

-------> Then what the established theory and BJT CE characteristics and load-line curves tell us about?

So far the explanation Tolga has stated for MOSFET will really help understand as how MOSFET works internally. Nice explanation!

Coming back to the question that what Sayeed has asked -

The concept of "CONTROLLED DEVICE" should mean "by which factor or input quantity the output quantity is controlled.

Thus a MOSFET is a voltage controlled device because its the voltage Vgs at the GATE you apply will turn ON or OFF the device and hence provide control over Id.

In case of a BJT, it is a current controlled device since the base current Ib governs the output collector current Ic.

Thank you all for sharing your inputs!

My answer is a little late, but still.

FETs can probably really be defined as the VCCS. An input current is really small.

BJTs - It is in fact a power controlled current source. We need input voltage and input current too - we need input power. An ideal CCCS must have a null intput voltage, thus null input power.  Take it easy,

Josef

Very precise answer...

... but to be extremely precise:) we should say that the separate FET (BJT) is not yet a current source... it is only a voltage-controlled current-stable resistor... ... It can be considered as a current source when combined with the power supply...

Hi Cyril,

of course it is not an ideal current source - but practically yes :-))

Josef

I mean it is not a current source but only a component of a current source:)

Quote Josef: BJTs - It is in fact a power controlled current source. We need input voltage and input current too - we need input power

Sorry, I am a bit late with my comment, nevertheless:

I think, it is NOT true that we need a base current.  Of course, there is a current and we cannot avoid it (unfortunately), however, the output current of the BJT clearly is controlled by the voltage Vbe only. The known relation Ic=f(Vbe) must not be interpreted as a cause-effect relationship.

But any Vbe needs some current (base) - see diode B-E.

Josef

Josef - what means "need"?

There is a current IB - yes, unfortunately. But it has no control or steering function. It is the voltage VBE only that causes the charged carriers to leave the emitter node with enough energy to "work" against the diffusion voltage. And then, the collector potential takes over. We do not "need" the current IB at all.  But it does exist without any doubt.

To "need" something means that the needed item should serve a specific purpose. What could be the puprose (or the task) of the base current  IB?

Hi Lutz, see below

"However,

the junction transistor does require finite base currents

in order to modulate the emitter and collector currents.

One of these sources of base current is the recombination

of electrons with holes in the base layer. This is unimportant

in many cases, and we shall neglect it as mentioned

above. The other source of base current consists

of a flow of holes from the base into the emitter. This

current is an ordinary junction current and can be calculated

by the methods described in Section 5. The

ratio of this hole current to the electron current can be

shown to be

eqv. (53)"

Hi Josef - indeed, the text in the document you have provided is somewhat misleading. (By the way - this conribution from Shockley is known to me since many years). Let me explain:

In Fig. 22(b), 22 (c) and the corresponding text it is described how the flow of electrons depends on the bias.  Consequently, Shockley states that "if it were true that....there would be no change in base current ....the transistor would behave much like an ideal negative grid triode" .

This bias-controlled electron current is the dominating quantity. However, because a change of the bias condition will also result in a small change of the base current (Shockleys term: "hole current") this hole current will have small effect on the total (electron) current through the device (according to Shockley this effect "modulates" the total current).

Finally, we read: "When the base potential is altered there is a change in electron current flowing ....to collector and a change in hole current from base to emitter."

Hence, it is clear that the base potential controls the collector (electron) current - with a slight influence of the base (hole) current.

Finally, here is a statement from the famous Barrie Gilbert:

BJT is a voltage-controlled current-source; the base current is purely incidental (it is best viewed as a „defect“)