Why isn't BJT voltage controlled and current controlled???

This might sound silly for some of you..
Many are debating over the fact whether BJT is current controlled or voltage controlled..

My doubt is this..
Voltage and current are related by the Ohm's law V=IR.. then V(be) should be proportional to I(b).. which means
The collector current I(c) should be proportional to both V(be) and I(b)

Why is this not true...

Because of the nonlinearity of the BE diode.

bjt is only a voltage controlled device not at all current control device the base emitter voltage is the voltage which control the amount of current into the collector .but you might argue when voltage and current are related by ohm law why it is not possible . it is possible but when you consider the blackbox view that is you are not interested in internal working of the transistor . this ohm law ignore the internal physics. so bjt is only a voltage control device which can turn off and on by vbe and not current .moreover it is the voltage which causes current and current cannot cause voltage.

take a look at this
https://www.edaboard.com/threads/196972/#post827199

Alex

alexan_e said:

take a look at this
https://www.edaboard.com/threads/196972/#post827199
Alex

Click to expand...

Hi Alex,

I know your way to argue and your final opinion regarding bjt control
However, for my opinion you make one fatal error: Some test with or without resistors that act as current-voltage transformers cannot reveal the truth. You treat the BJT as a 3-terminal black box - without regard to the semiconductor physics. This approach may be helpful to derive or to understand some describing formulas.
But it cannot answer the principle question: Current or voltage controlled?
There is no way out: You must respect some physical basics and then you soon will arrive at William Shockley who described first the relationship between voltage and current across a pn junction with his classical formula :

I=Io*[exp(Uj/Ut) -1].

In words: The current through a forward biased pn junction depends on the voltage Uj across the junction.
What do you need else?

That equation describes a relation but you can solve it for
either voltage or current as the dependent variable.

The only gut-reasonable explanation I've ever seen for
BJT operation, is charge and time (current) based.

In useful regions of operation, only the current mode
is anything like linear.

If somebody is looking for a pat on the head for being
"right", that's very different from being a useful design
approximation. You can estimate Ic pretty reasonably
from Ib or vice versa. You can't get much of anything
from Vbe without a whole bunch of other detail and
calculation.

Hi dick_freebird,

according to my understanding the original question was and is related to semiconductor physics: What physical phenomenon controls the collector current?

Quote: That equation describes a relation but you can solve it for
either voltage or current as the dependent variable.

Yes, mathematically you can solve each equation with - let's say - 5 parameters to each of these five.
But this has nothing to do with the "cause and effect" question. It`s pure formalism.

Quote: The only gut-reasonable explanation I've ever seen for
BJT operation, is charge and time (current) based.

Charges can move only if a voltage (field) is existent.

Quote: In useful regions of operation, only the current mode
is anything like linear.

This is irrelevant with respect to the point under discussion

Quote: If somebody is looking for a pat on the head for being
"right", that's very different from being a useful design
approximation. You can estimate Ic pretty reasonably from Ib or vice versa.
You can't get much of anything from Vbe without a whole bunch of other detail and
calculation.

This also was not the question.
A "useful" approximation may be - from time to time - a good tool.
However, it is not bad to know the physical background for such approximations.
Only in this case you are able to decide if (and under which conditions) such a simplified view is appropriate.
Question: Can you explain the function principle of the current mirror using your current-only oriented approach?
And what about transistors working in the log-domain?

Regards
LvW

LvW said:

What physical phenomenon controls the collector current?

Click to expand...

I always believe that understanding the physical process inside the semiconductor is ke KEY to demystifi transistor action.

XT.5155 said:

I always believe that understanding the physical process inside the semiconductor is ke KEY to demystifi transistor action.

Click to expand...

More than that: It is also the key to understand the operation of the whole circuits, the role of each part within the circuit and - last not least - it provides the basis to develop novel applications. The knowledge of some rule of thumbs certainly is not enough.

QUOTE=LvW Hi dick_freebird,

according to my understanding the original question was and is related to semiconductor physics: What physical phenomenon controls the collector current?


Carriers injected into the base as forward B-E current, violate
charge neutrality and the C-B barrier is lowered until the charge is
compensated by opposite charged carriers from the collector.

The carriers from the collector transit the base quickly spending only
Tf there. The minority carriers in the base have a lifetime Ts. Thus a
single base charge is compensated (during its lifetime) by Ts/Tf
carriers transiting from C to E across the base. This is the mechanism
that sets up your peak hFE.

Now the base charge is not static, it decays and must be replenished
as Qbase/Ts = Ib (leaving aside other losses). A fixed Ib induces a fixed
Qb which is compensated by Ic*Tf.

Or something like that. I got my B+ and that was the last time I used it
for anything.

I'm sure the intervening years have produced more sophisticated theory
but this one I found reasonable.


Question: Can you explain the function principle of the current mirror using your current-only oriented approach?
And what about transistors working in the log-domain?

Sure, I'll take a stab. The principle is that for equal current density in the
emitter there will be an equal Vbe for an equal device. And vice versa.
This in turn means that in a rack of identically made, equipotential E-B
junctions the base current will divide according to area. Collector current
follows simply. The reference collector subtracts as much of the reference
current as needed until the base is starved of the base current it needs
to do so, the current still divides evenly by area and the negative feedback
of the collector in the end sets the final base current left to distribute.

Whether voltage causes current or current causes voltage is immaterial,
the principle is only that equality of construction and connection produces
equality of conduction.

The principle is that for equal current density in the
emitter there will be an equal Vbe for an equal device.

Hi dick-freebird,
thanks for your explanation. I could follow it - and I agree. But, interestingly, your explanation starts with the equality of Vbe.
For my opinion, that answers the question already.

Whether voltage causes current or current causes voltage is immaterial,

I think, a current cannot "cause" a voltage. (On the other hand: Perhaps there are some exceptions? Some sensors?)
For example, the voltage divider rule according to which the current through both parts "generates" voltages (depending on the resistor ratio) is only an artificial tool which helps to arrive at a simple conception and simple formulas.
I agree with you that - in principle - the question quoted above is "immaterial". However, in context with the original question (which asks for the physical background) it has come automatically "on the table".

Regards
LvW

I think, a current cannot "cause" a voltage.

Click to expand...

I can't follow this consideration, in my view, there's a duality of current and voltage, each can "cause" the other in a specific situation. A current source isn't just an abstraction, it can exist as a standalone device without requiring a driving voltage. Consider e.g. a ribbon generator, that primarly generates a current of charge carriers. In so far, something like claiming voltage superior to current can't answer the original question.

I completely agree with "transistor is voltage controlled" in terms of modelling. Nevertheless I will often treat a transistor as a current gain element in intuitive circuit analysis and not think about voltage control when looking at a single current mirror.

For now if you assume that, the BJT is current controlled device, then how Ib is able to control Ic. because Ic is in milli amps whereas Ib is in microamps ..

Thanks,
SP3

FvM said:

I can't follow this consideration, in my view, there's a duality of current and voltage, each can "cause" the other in a specific situation. A current source isn't just an abstraction, it can exist as a standalone device without requiring a driving voltage. Consider e.g. a ribbon generator, that primarly generates a current of charge carriers. In so far, something like claiming voltage superior to current can't answer the original question.
.....

Click to expand...

Ha, I know that my wording (a current cannot cause a voltage) was a bit "provocative" - and as you can read I was not absolutely sure about that (some exceptions?). And, of course, I did expect an answer like yours.
On the other hand, assuming there is a "current source" as a standalone device I ask myself: Which effect causes the carrier (charges) to move and to form this phenomenon that we call "current"? Isn't an electrical field caused by a potential difference (voltage) necessary?
I must confess, I really tend to believe that the term "current" really is - more or less - an abstraction that helps to describe some effects that are observed if we bring a voltage source with some conducting materials in contact (as you can see: I avoid the term "current").
In this context, I remember the abstraction we are using by explaining the BJT principle (movement of electrons and holes).
OK, they move - but with a velocity that is much, much lower than the velocity of the resulting "current". But it's a nice conception, no doubt about it.
I am afraid, we are moving into the region of nuclear physics (and philosophy?).
Therefore, I come to and end.
Regards
LvW

---------- Post added at 15:32 ---------- Previous post was at 15:29 ----------

sp3 said:

For now if you assume that, the BJT is current controlled device, then how Ib is able to control Ic. because Ic is in milli amps whereas Ib is in microamps ..
Thanks,
SP3

Click to expand...

This is to be answered from somebody who thinks that it is current-controlled.

I am afraid, we are moving into the region of nuclear physics (and philosophy?).
Therefore, I come to and end.

Click to expand...

Yes. I also don't think, that we need to move here to discuss the original question. I went on this excursion to explain, why I think the statement "current cannot cause a voltage" brings us nowhere in terms of physics. Of course, there may be different opinions how to describe the respective physical processes.

For now if you assume that, the BJT is current controlled device, then how Ib is able to control Ic. because Ic is in milli amps whereas Ib is in microamps ..

Click to expand...

Following this - excuse me - simple-minded argument, how do you explain that a voltage controlled device exposes voltage gain? Neither current control nor voltage control involves any magnitude restrictions, I think.

from the subject the question is y? isnt the bjt voltage controlled
- cannot adj. the voltage across the pn junction. it will always stay 0.6 for Si.
- reasonin i keep in my mind is that the only way to increase the conduction of the O/P of the bjt is to increase the conduction of the I/P of the bjt. the diode is a very selfish device and will allow the pn junction to be permanently damaged by openning fully. this will pull large amounts of current Ic. and burn the device. to stop this we limit the resistor Rb. compromising how much we can really drive the O/P... V and I exist at the same time. correct.
-but in the case of the bjt. the amount of current flowing I/P will comtrol the amount of Ic
-fet.. the amount of voltage across the G to S will control the amount of current Id
-thus thats why we say current controlled device.

cannot adj. the voltage across the pn junction. it will always stay 0.6 for Si

Click to expand...

No it doesn't stay 0.6. It's around 0.6 in active operation. Strictly spoken it follows the Shockley equation as explained by LvW in pot #5. I=Io*[exp(Uj/Ut) -1] The difference matters.