A simple transistor question

[TimFox]

As a simple explanation to the novice, it is easier to start with the diode equation, have the base-emitter voltage produce an emitter current, then state that the higher voltage on the collector attracts most of that current.
Beta varies all over the map between devices with a given part number, and over a reasonably wide range with respect to emitter current, while the transconductance from the Schockley equation is better defined.
For DC, the voltage and current happen together, just like in a resistor, but not linearly related.
Otherwise, you need to explain how a low current introduced into the base terminal magically induces a larger current in the collector lead.
Of course, then the student should learn the more detailed equations for the DC performance, and then proceed to the dynamic performance.

[Sredni]

The explanation of how a small base current can draw a large collect current is given on Streetman, for example.
On a less device oriented textbook, Sedra and Smith explain how a diode works by driving it with an ideal current generator and how that current ends up changing Vbe, which in turn...
It's a chicken and egg problem.
But as an agnostic, I don't want to start a religion war here.

[T3sl4co1l]

Ah, EESE?  Being a community, like any other, it could simply be you'd get different answers at different times of the day or week -- different users online.  You're as likely to get flamed here (or just get low quality or off topic comments) as useful comments; the only difference is they have tools available to the users to self-moderate.

Being a didactic question -- it could well be better placed on https://academia.stackexchange.com/ than EE.  You run the risk of subject matter experts not being present -- but perhaps the meat of the question is recognizable enough, or the subject can be explained well enough to them, to figure out a solution to the prompt.

I guess I'd be surprised if Meta didn't suggest this, but maybe there aren't as many people there, or, that care.

Tim

[magic]

I have to disagree with this. Voltage and current, like electric and magnetic field are concomitant manifestations of a single entity. There is not a cause and an effect (like in an EM wave it's not the change in magnetic field that causes the change in electric field and so on).
One can choose to see a BJT as voltage controlled or as current controlled, or even as charge controlled.

One can choose anything, but here's something:
You can write a reliable equation for emitter current as a function of base-emitter voltage, or vice-versa, which works for any collector voltage and temperature.
There is no straightforward and equally general equation for emitter current as a function of base current. Nor for collector current.

Then there is the fact that - as you said in another message - it might be easier to see the transistor as current controlled because the voltage would be too sensitive to fiddle with, but fundamentally, one can explain how a BJT works by injecting current

Well, yes, because the base is a feedback output which gives you approximately 1% of collector current, at least under some conditions, and for some value of "approximately"

As a simple explanation to the novice, it is easier to start with the diode equation, have the base-emitter voltage produce an emitter current, then state that the higher voltage on the collector attracts most of that current.

This doesn't seem to account for saturation.

Otherwise, you need to explain how a low current introduced into the base terminal magically induces a larger current in the collector lead.

Yep. The magic involves forward biasing the junction as a result of pulling current from it, and we are back to voltage.


BTW, the original question was extremely low quality and a clear sign of somebody being not even close to understanding fundamentals. First, the use of Ohm's law to talk about diodes. Second, the confusion/errors in the definition V2. Third, why should anyone care about Vce being greater than Vbe? What it was that the guy even wanted to know?

The only sensible answer to such question is a full, ground-up introduction to BJT theory. You can become the 1000th person to write one if you think there is something novel you can add to the topic, or just tell the OP to RTFtextbook. It appears that the general attitude at SE is to do the latter in such cases.

[aneevuser]

I have to disagree with this. Voltage and current, like electric and magnetic field are concomitant manifestations of a single entity. There is not a cause and an effect (like in an EM wave it's not the change in magnetic field that causes the change in electric field and so on).
One can choose to see a BJT as voltage controlled or as current controlled, or even as charge controlled.

One can choose anything, but here's something:
You can write a reliable equation for emitter current as a function of base-emitter voltage, or vice-versa, which works for any collector voltage and temperature.
There is no straightforward and equally general equation for emitter current as a function of base current. Nor for collector current.

Yes, indeed. And that is of course because there are well understood physical principles which relate BE voltage to emitter current (Boltzmann distribution), but there aren't any such principles which relate base current to emitter current - that depends on the details of device manufacture.

And it seems to me that it's fundamentally silly to promote a BJT as base-current controlled, when circuits generally do not (or never?) "control" base current. Usually, a BJT is controlled by choosing a DC operating *voltage* on the base, and a small signal *voltage* is then applied to the base, and the equations which predict the linearised behaviour of the BJT are written in terms of voltage. It's voltage all the way, no?

If we were controlling base current, wouldn't we be seeing programmable current sources attached to the base?

[TimFox]

The vacuum triode, operated in the normal space-charge-limited regime, is always considered to be voltage actuated (i.e., the cathode current is a function of the grid-cathode and plate-cathode voltages).
If you forward-bias the grid-cathode diode, the same space-charge equations give you the cathode current, but now some of the cathode current flows to the positive grid.
It is still considered voltage actuated, and the ratio of cathode current to grid current is not considered a control parameter, but rather as a result of the positive grid voltage.

[rfeecs]

It's amusing that Shockley wrote that in both the vacuum tube and the bipolar transistor, "one current is controlled by another":

From "BSTJ 28: 3. July 1949: The Theory of p-n Junctions in Semiconductors and p-n Junction Transistors. (Shockley, W.)"
https://archive.org/details/bstj28-3-435

[rfeecs]

Yes, indeed. And that is of course because there are well understood physical principles which relate BE voltage to emitter current (Boltzmann distribution), but there aren't any such principles which relate base current to emitter current - that depends on the details of device manufacture.

 

[TimFox]

The Schockley article referencing grid current reminds me of an advanced topic in vacuum-tube electronics:  "charge control", as discussed in one of the last great textbooks to include tubes, "Amplifying Devices and Low-Pass Amplifier Design" by E M Cherry and D E Hooper, Wiley 1968.  This concept is useful in discussing high-frequency behavior in tubes, but is not normally taught as the first item in freshman vacuum-tube electronics.   Instead, one usually started with the Child-Langmuir Law, where the current in a space-charge limited thermionic diode is proportional to V^3/2, and then proceeded to add a grid and calculate the field at the cathode due to grid and plate voltage, introducing the parameter [mu] to represent the effect of the grid on the cathode field due to the plate voltage.

[aneevuser]

Yes, indeed. And that is of course because there are well understood physical principles which relate BE voltage to emitter current (Boltzmann distribution), but there aren't any such principles which relate base current to emitter current - that depends on the details of device manufacture.

 

Yeah, harsh. But fair.  Good point well made, and all that kind of stuff. The wording after the "but" is indeed unfortunate - let me reword it to something like "the physical principles which determine it are more easily influenced by device manufacture than those that determine the relationship between V_BE and I_E."

Happier?

[rfeecs]

"the physical principles which determine it are more easily influenced by device manufacture than those that determine the relationship between V_BE and I_E."

Looking at a textbook:
https://www.chu.berkeley.edu/wp-content/uploads/2020/01/Chenming-Hu_ch8-2.pdf

The equation for collector current (8.2.7) is exactly the same form as the equation for base current (8.3.3).  The equation for beta (8.4.5) depends on the same physical parameters as the currents: diffusion lengths, base and emitter widths, doping concentrations and intrinsic carrier concentrations.

They depend on exactly the same physical principles.

Then again, almost all the textbooks say something like "if the potential across the junction is changed, current flows ... " 

And most people tend to look at it that way.  That doesn't mean it is the only way to look at it.

[magic]

Then again, almost all the textbooks say something like "if the potential across the junction is changed, current flows ... " 

And most people tend to look at it that way.

Perhaps something to do with the "Vbe" appearing in your equations.

I wonder where is the "small currents control large currents" explanation of bandgap voltage references or, hell, even current mirrors.
 

[rfeecs]

I wonder where is the "small currents control large currents" explanation of bandgap voltage references or, hell, even current mirrors.
 

You're right, it's not "current controlled".  It's temperature controlled!   

[Sredni]

Ah, EESE?  Being a community, like any other, it could simply be you'd get different answers at different times of the day or week -- different users online.  You're as likely to get flamed here (or just get low quality or off topic comments) as useful comments; the only difference is they have tools available to the users to self-moderate.

This is not the first time I have witnessed the closing of potentially very interesting question out of the ignorance of those who voted to close them, and asking why a question was closed is like being appointed defense attorney in a witch trial :-) . The verdict can't be wrong, mods/curators side with each other, and the reasons given (if any) are opaque to say the least.
Arbitrariness and unaccountability. I don't think even giving them a bodycam would change anything. :-)

The youthful porpoise of EESE should not be that to boost one's ego and/or punishing seemingly lazy users. You will do no good to the site by deleting their valid questions because they did not format the text to your liking, or they did not adopt your notation.  Nor will you succedd in educating them when they do not immediately reply to comments (because, well, the Earth is not flat), or even when they are just plain lazy. For the simple reason that even when they are in the wrong either they just don't learn or they are never the same student: they keep changing every time. Therefore, if the purpose is to provide answers to the community, fix the godd***n question, not for the lazy ba**ard, but for all other users that can take advantage from the ensuing answers (this is how the site generates traffic, after all).
But of course, one needs to understand the question or put their ego apart to type a space here and there, instead of voting to close it. This is especially damaging to the site when users who do not care about 'punishing' people for their bad formatting or for speaking English as a second language, might have started working on the question, to only find it closed when they log in to answer.

From time to time I test the site to see if something has changed and would make it worth it to go back, but they seem unable to evolve. They lost me a long ago as a contributor (I make exceptions to maintain the answers I gave on the Lewin ring and for some rare occasion when I stumble on something particularly irking) but I really wished they did not drag the level down too much, tho.

So many potentially interesting questions wasted to show off their 'power to close'. It's kind of sad.
(On the bright side, if you need to select a resistor to light an LED...)

I guess I'd be surprised if Meta didn't suggest this, but maybe there aren't as many people there, or, that care.
Tim

Yeah, this is the impression I received: nobody really cares. They seem to be there for brownie points.

[Sredni]

BTW, the original question was extremely low quality and a clear sign of somebody being not even close to understanding fundamentals. First, the use of Ohm's law to talk about diodes. Second, the confusion/errors in the definition V2. Third, why should anyone care about Vce being greater than Vbe? What it was that the guy even wanted to know?

See? This is what I mean when I say that they close potentially interesting questions out of their own ignorance.
There is this vicious rumor that a diode in not a resistor. But it is. It is a resistor in that its state is described by the instantaneous values of voltage and current and not their derivatives (or integrals). Being nonlinear, the device resistance is a function of the current - or the voltage (I am still an agnostic). And I am not talking of the differential resistance, but to full blown resistance computed as the ratio of the voltage across the diode and the current flowing through it. It all depends on how this student has been exposed to the concept in class: not all the world is alike.

And this brings us to another interesting point that could have been pointed out in an answer to that question: the fact that a transistor does not only amplify 'differential signals' about the biasing point (as was suggested by one of the comments). In one of the examples I gave, an high current but low voltage supply can be used to drive a high voltage load. This is using the transistor as an amplifier: 3.3V input, 10.some V output. A common collector can be used to amplify current (not the differential signal, the whole shebang) and a common emitter can do a little bit of both.

And no, you do not need to do a full course on transistors to answer the question. I will edit my answer by formatting the unnecessary details in a small font (just that) - you do not need that, but I felt it would have made the answer more general.
All this student was asking was: if Rpn<Rnp and IE>IC how can I say Rpn*IE < Rnp*IC? And the answer is: because IE is basically the same as IC. Yet, the real explanation is that of how IC can be essentialy equal to IE no matter how I choose VCC in the battery-only circuit or the load resistor in the resistor-laden circuit.
And that can be explained by the fact that in a reverse biased diode the characteristic is flat (the current is essentially identical to the reverse saturation current, no matter what the reverse voltage). Personally, this is what made me really understand how BJT works.

(I cannot help but notice that nobody pointed out that in one of my examples with added resistors I get |VCB| < VEB :-), the reason this happens in the second type of circuit  can be shown graphically by introducing the load lines.)

[Sredni]

I have to disagree with this. Voltage and current, like electric and magnetic field are concomitant manifestations of a single entity. There is not a cause and an effect (like in an EM wave it's not the change in magnetic field that causes the change in electric field and so on).
One can choose to see a BJT as voltage controlled or as current controlled, or even as charge controlled.

One can choose anything, but here's something:
You can write a reliable equation for emitter current as a function of base-emitter voltage, or vice-versa, which works for any collector voltage and temperature.
There is no straightforward and equally general equation for emitter current as a function of base current. Nor for collector current.

Yes, indeed. And that is of course because there are well understood physical principles which relate BE voltage to emitter current (Boltzmann distribution), but there aren't any such principles which relate base current to emitter current - that depends on the details of device manufacture.

The body of relationships relating transistor parameters to voltage is the consequence of the framework into which solid state theory has evolved. Everything is reconduced to energy levels, so everything electronic seem to be caused by electric potential and voltage.
But you cannot have a voltage without displacing charge - first in your battery/generator and then at your device's terminals. This is at first a matter of surface charge, but then it's the charge flowing inside the conductors that allows you to mantain the surface charge in a configuration that will sustain the voltage. It's a chicken and egg situation. And even in the device, if you want to change the thickness of the depletion region, and hence the voltage that opposes the flow of charge, you need to supply charge or your voltage won't change (see the elementary treatment of the diode in Sedra & Smith, for example). Heck, the diode's built-in voltage itself the result of charge movement initiated by difference in concentration.
Even the MOSFET requires a lot of current during transients, because if you want to change the voltage (in what is fundamentally a capacitor) you cannot do that without moving charge.

If anything, the model that better describes electronic devices is the charge transfer model: you move charge from here to there (there is a current) and at the same time the potential in space changes. The potential is just a manifestation of the position of charges in a given configuration. Yes, if you succeed into fixing the charges in that position, from then on you can have a potential field without current (electrostatics) but to reach that configuration (in an empty space with charges at infinity) you need to move charges first. And the potential in all space will change at the same time as your configuration of charges change (actually, the information on the new charge's position will propagate at the speed of light.)
Then, once you have reached your configuration (of charge and potential), if you want to change the potential in space, you need to add, remove and in any case move charges. Again the potential will change at the same time the charges change configuration.

And it seems to me that it's fundamentally silly to promote a BJT as base-current controlled, when circuits generally do not (or never?) "control" base current. Usually, a BJT is controlled by choosing a DC operating *voltage* on the base, and a small signal *voltage* is then applied to the base, and the equations which predict the linearised behaviour of the BJT are written in terms of voltage. It's voltage all the way, no?

If we were controlling base current, wouldn't we be seeing programmable current sources attached to the base?

How do you think I set the collector current in my two examples? By choosing VEB? No - I programmed IE by choosing a resistor of value (VEE-0.7V)/IE (this is by modeling the input diode as a vertical line at 0.7V).

[bdunham7]

There is this vicious rumor that a diode in not a resistor. But it is. It is a resistor in that its state is described by the instantaneous values of voltage and current and not their derivatives (or integrals). Being nonlinear, the device resistance is a function of the current - or the voltage (I am still an agnostic).

Yeah, sometimes when people refer to resistance or impedance (zener impedance for example) they mean dV/dI, not V/I.  So there's some ambiguity there.  But to be called a resistor and have the term mean anything at all, I think it is implied that V/I = dV/dI or at least is very close.  Another way of saying it is calling it 'ohmic'.  Calling any device a 'resistor' just because there is a V/I number at any given point reduces the term to meaninglessness.  IMO, of course.

[Sredni]

Well, you won't find carbon powder or thin metallic film inside, but as far as the VI characteristic go, that is a nonlinear resistor from the point of view of circuit theory.

Of course when you consider second order effects, you will have all kind of parasitics, starting from the two capacitances. But if we work at DC, as specified from the original circuit...

[bostonman]

https://www.chu.berkeley.edu/wp-content/uploads/2020/01/Chenming-Hu_ch8-2.pdf

Which book is that from? I tried working backwards from the hyperlink, but couldn't get to the source.

As for where this topic has gone and the disagreements, as with many topics, people understand things differently. As an example, maybe it's my basic knowledge of BJTs and MOSFETs that keeps my questions and interpretations basic.

With both a BJT and MOSFET, I look at it as the base/gate starting the whole process. Neither one will turn on unless voltage and enough current is on the base/gate. At that point it's just ohms law. If the BJT is turned on fully, the DC voltage on the collector is 10v, then (with a perfect transistor) Ic would be 10v / Rc. If Ic = Ie, then it's Ic/RE for the voltage.

In the case of a MOSFET, it's basically (in my head) similar steps. Hence, this is why I never grasped why a BJT is a current amplifier and a MOSFET is a voltage amplifier.

Due to seldom needing to calculate transistor circuits, sometimes I get confused when a resistor is on the emitter.

With a BJT and the emitter tied to ground, the base is (with a perfect transistor) 0.7v. If a resistor is on the emitter to ground, then that changes the base voltage which changes the base current, which changes the collector current because beta changes, etc... Now I'm confused.

On a side note, at my old job, we use to give basic electronics tests to applicants. My boss would give a quick overview of the test and I'd over hear him say that the resistor on the emitter and the resistor going from base to ground are in parallel and the circuit should be easy to calculate. I use to laugh because he moved up the corporate ladder by making upper management believe he was some great engineer who ran large groups at previous jobs.

[T3sl4co1l]

There is this vicious rumor that a diode in not a resistor. But it is. It is a resistor in that its state is described by the instantaneous values of voltage and current and not their derivatives (or integrals). Being nonlinear, the device resistance is a function of the current - or the voltage (I am still an agnostic).

Yeah, sometimes when people refer to resistance or impedance (zener impedance for example) they mean dV/dI, not V/I.  So there's some ambiguity there.  But to be called a resistor and have the term mean anything at all, I think it is implied that V/I = dV/dI or at least is very close.  Another way of saying it is calling it 'ohmic'.  Calling any device a 'resistor' just because there is a V/I number at any given point reduces the term to meaninglessness.  IMO, of course.

Right -- the whole point of highlighting resistance, it being this whole big thing, big enough to call it a "law" -- is that it's this, perhaps unexpectedly, near perfectly linear (very flat, over a very wide range of scales) characteristic happens to show up, and so we celebrate that simplicity, for it makes our jobs so very much easier.

And so anything that is significantly nonlinear, is most definitely not a resistor, at least not any good of one.  So It leads me to wonder where the student even got this notion in the first place, did they come up with this themselves?  What is OP's relationship here, are they tutoring?  Is this entirely hypothetical (fabricated)?  If so, is it even a realistic scenario -- is it based on some other real experience?

In short, it strikes me that, it was likely someone's duty to inform the student that, no, in fact transistors are NOT resistors, they cannot be treated that way, and it seems like OP is stretching this idea much farther than it can go.

So I guess I'd be inclined towards the self-learning angle, that seems the most reasonable origin.  But OP has no duty to entertain this attempted [mis]application of theory; I'm all for, not disregarding ones' ideas outright, but acknowledging it should be enough, and then immediately moving on to why it doesn't work, and how ones' perspective can be shifted to a more practical direction.

(Again, usual caveats apply: I'm very far from an expert on teaching theory; no offense intended, just being my usual frank self, trying to get at the core of things here.)

Tim

[Sredni]

Right -- the whole point of highlighting resistance, it being this whole big thing, big enough to call it a "law" -- is that it's this, perhaps unexpectedly, near perfectly linear (very flat, over a very wide range of scales) characteristic happens to show up, and so we celebrate that simplicity, for it makes our jobs so very much easier.

And so anything that is significantly nonlinear, is most definitely not a resistor, at least not any good of one.

Leon O. Chua, Charles A. Desoer, Ernest S. Kuh
Linear and Nonlinear Circuits
1987, McGraw Hill

I would call this THE bible of circuit theory.
p. 50 : section 1.2: The nonlinear resistor
the first example is the ideal diode.


For BostonMan:

Chenming C. Hu
Modern Semiconductor Devices for Integrated Circuits
2009, Pearson
384 pp.
 (pdf version on author's website https://people.eecs.berkeley.edu/~hu/ )

Modern Semiconductor Devices for Integrated Circuits, First Edition introduces readers to the world of modern semiconductor devices with an emphasis on integrated circuit applications.
Written by an experienced teacher, researcher, and expert in industry practices, this succinct and forward-looking text is appropriate for anyone interested in semiconductor devices for integrated curcuits, and serves as a suitable reference text for practicing engineers.

[T3sl4co1l]

Right -- the whole point of highlighting resistance, it being this whole big thing, big enough to call it a "law" -- is that it's this, perhaps unexpectedly, near perfectly linear (very flat, over a very wide range of scales) characteristic happens to show up, and so we celebrate that simplicity, for it makes our jobs so very much easier.

And so anything that is significantly nonlinear, is most definitely not a resistor, at least not any good of one.

Leon O. Chua, Charles A. Desoer, Ernest S. Kuh
Linear and Nonlinear Circuits
1987, McGraw Hill

I would call this THE bible of circuit theory.
p. 50 : section 1.2: The nonlinear resistor
the first example is the ideal diode.

I don't have this book handy, sorry.  Are you saying that the motivation for analyzing it as a "resistor" comes from this reference then?

Tim

[Zero999]

c) to understand that the base-current-to-collector-current relationship/beta is more like a correlation than a causation - collector current is "caused" by the V_{BE}, but correlates well with the base current.

The extent to which that can be put across to any given student will vary, of course.

[prediction: this thread will gradually devolve into the standard heated argument as to whether BJTs are current-controlled or voltage-controlled]

You cannot show that voltage causes current or that current causes voltage.  You have a relationship, not a cause/effect.  So better not to bring it up. 

https://www.eevblog.com/forum/beginners/transitor-the-base-pin/https://www.eevblog.com/forum/projects/_n-channel_-vs-_npn_-name-confusion/   

[TimFox]

At best, Ohm's Law is an accurate approximation to the current vs. voltage in a two-lead passive component that we call a resistor.
Even resistors found at Mouser, such as high-megohm thick-film short surface-mount parts (e.g., 50 megohm 0603 package) have measurable non-linearity.
But, if we are to confuse new students by calling any two-terminal passive component a "resistor", do we stop before we hit the NE-2 neon bulb?
There is a well-worked-out theory of non-linear components, including resistors and capacitors, but I don't believe it treats charge storage in a PN diode.

[magic]

I would like to propose the radical notion that you are wrong and it actually is resistors which are a subset of diodes, namely those diodes with variable saturation current which satisfy:

Is = e^{(ln(Vf)-Vf/Vt+X)}

where Vf is the forward voltage, Vt is the thermal voltage and X an arbitrary constant which characterizes the particular specimen of resistor diode.


In other words, you can always fit any model to a single operating point. But the result is not guaranteed to be helpful in any way or worth caring about.

Talking about static "resistance" of a collector is even more meaningless, because Vce and Ic are two almost completely independent variables and can be anything without affecting the other much. Their static ratio truly tells you nothing at all about anything.