transitor: the base pin.

[LvW]

OK gentlemen - I resign myself.

I still think that we are in „A Free & Open Forum For Electronics Enthusiasts & Professionals“ but I don`t like to further respond to „arguments“ like   

 * nonsense.
 * [misunderstandings] Yes, exclusively on your part.
 * ...quoting irrelevant facts or terminologies serve no purpose in a discussion other than to undermine your own credibility, especially when such quotes showed your   complete misunderstanding of the quoted items, or your "facts" are erroneous.
* It clarifies nothing other than your stubbornness!

I am in the age of  73 - and I still feel able and willing to learn.
But I am not sure if this is true for all of you who have answered as quoted above.
I came to this conclusion because - up to now - nobody was able or willing to respond technically to the arguments and examples  I have mentioned in favour of voltage control (Re-feedback, dVbe/dT=-2mV/K, class-B biasing,...).
If I am totally wrong - are you really not in the position to show (with objective technical arguments!) where/why I am wrong instead of starting personal attacks?
 (But I can imagine what`s behind all this).

Final remark (surprising observation):
In my life I had, of course, many discussions about technical details - in most cases on a fair and  respectful basis.
However, as soon as the question is discussed how the BJT is controlled (because of conflicting explanations in different books/articles) very often there is a surprising change:
Some people stop exchanging technical arguments and treat this matter as a kind of „religion“ - claiming that they have learned decades ago the current-control property and - hence - this would be the truth. Full stop. Without any justification and without any counter argument against voltage-control.
Interesting observation.   

[Simon]

If you talk amplifier circuits then yes you talk voltage because the feedback system uses voltage. but the transistor is a current controlled device. I have never heard it described otherwise. As soon as you stick a resistor in a BJT circuit you control the circuit (not the BJT) with voltage.

[kxenos]

This thread has become so embarrassing to read. You can't talk to a 73-old university professor like that. It doesn't matter if he's right or wrong, there is a better way to respond. I'm out of this thread

[dannyf]

I still think that we are in „A Free & Open Forum For Electronics Enthusiasts & Professionals“

It is, except that it requires participants willing and able to listen. You have demonstrated neither.

I came to this conclusion because - up to now - nobody was able or willing to respond technically to the arguments and examples  I have mentioned in favour of voltage control (Re-feedback, dVbe/dT=-2mV/K, class-B biasing,...).

Those are examples that have zero bearing on if a bjt is current or voltage controlled, a theme that has been repeated multiple times for you.

Take the tempco question for example, contrary to popular believes, it is NOT a constant. We often use a constant to approximate it but in fact it varies with other factors, including temperature itself.

As to emitter / source resistor: it is always "current" feedback, regardless of if a current or voltage controlled device is used. IE., it would be current feedback even if a fet is used.

If I am totally wrong - are you really not in the position to show (with objective technical arguments!) where/why I am wrong instead of starting personal attacks?

You are totally wrong. Others have pointed out where and why. You don't realize that you are totally wrong because you are not willing to listen to the facts presented by the other side, and you insist on asking people to answer questions that have nothing to do with this discussion, which is what frustrated people in this discussion.

As to your tendency of calling for authority, that's a show of weakness. Your arguments should stand on their own. The fact that some authorities somewhere else had cited it has no bearing whatsoever in a "technical" discussion.

[LvW]

I have never heard it described otherwise.

Yes - and this is exactly the problem. Once and again, one should ask themselves if there is something to review.
Earlier in this thread I have mentioned specialists from Universities (Berkeley, Stanford,...) as well as the well-know "Art of Electronics".
I have offered corresponding references, but nobody was interested.
In contrary, it was me who was qualified as "stubborn".

[c4757p]

the transistor is a current controlled device. I have never heard it described otherwise.

Well that settles that, Simon hasn't heard it any other way.

For those of us whose heads aren't up our asses - there are multiple levels to the function of a BJT, and you can look at it on any of them. This is why everyone - including some quite knowledgeable people, as well as some quite ignorant ones - is disagreeing about this. Voltage at the B-E junction causes current through the device, most of which overshoots to the collector, blah blah... it depends on how you look at it.

Of course, let's not let that get in the way of a good argument.

[LvW]

As to emitter / source resistor: it is always "current" feedback, regardless of if a current or voltage controlled device is used. IE., it would be current feedback even if a fet is used.

Perhaps you should refresh your knowledge about feedback and realize in which cases the input resistance at the base/gate increases due to an emitter/source resistance.
Or do you even deny that this resistance goes up?

By the way: Interesting observation. The S-G path of a FET receives a feedback current. 
This statement (current feedback for FET`s) shows your qualification for this discussion.

[Simon]

the transistor is a current controlled device. I have never heard it described otherwise.

Well that settles that, Simon hasn't heard it any other way.

For those of us whose heads aren't up our asses - there are multiple levels to the function of a BJT, and you can look at it on any of them. This is why everyone - including some quite knowledgeable people, as well as some quite ignorant ones - is disagreeing about this. Voltage at the B-E junction causes current through the device, most of which overshoots to the collector, blah blah... it depends on how you look at it.

Of course, let's not let that get in the way of a good argument.

 

And we started with a very simple question begging a basic clarification and if I'm not mistaken we have gone to a whole new level a few levels above the normal one.

[Simon]

So the evidence presented so far is this wishy washy document : http://www.vanderbilt.edu/AnS/physics/brau/H182/Theory%20of%20Transistors.pdf that goes into very little detail.

If by voltage controlled it is meant that there has to be a voltage difference for anything to happen then that is a poor basis as we can argue the opposite that you can have all of the voltage you like but no current and of course still nothing will work in your model or real life.

If I have read right people are saying that the art of electronics supports "the voltage theory" well I've just skimmed through my copy and all formula's that involve voltage also involve resistance values in a circuit!

Of course voltage has a part to play but it's not strongly linked to BJT basic behaviour other than the obvious requirement for a voltage in order for current to flow.

[dannyf]

Or do you even deny that this resistance goes up?

It reminds me of "You want the truth? You can't handle the truth!" - that's a great movie.

You keep asking those tangential questions and insisting on others saying things they didn't say.

In your case, you want a "technical" discussion but you cannot handle a "technical" discussion. You don't have the ability to comprehend the technical details, nor do you have exhibited any willingness to understand what's being communicated to you.

[c4757p]

The voltage applied to the BE junction is what causes the current to flow; this is what people mean by voltage-controlled. Some consider the "control" to be what is applied to the pin; some consider it to be what the charge carriers do inside the silicon. Neither of them is wrong, until they start claiming everyone else is.

[LvW]

Or do you even deny that this resistance goes up?

It reminds me of "You want the truth? You can't handle the truth!" - that's a great movie.
You keep asking those tangential questions and insisting on others saying things they didn't say.
In your case, you want a "technical" discussion but you cannot handle a "technical" discussion. You don't have the ability to comprehend the technical details, nor do you have exhibited any willingness to understand what's being communicated to you.

Dear forum member dannyf., if you are fair you will have noticed that I have - earlier in the thread - in detail explained my example, but I have no problems to repeat it here again:

1.) First statement: The input resistance of an amplifier goes up if a voltage is fed back to the controlling inverting node and it goes down if a current is fed back. (These are basic rules from feedback theory).
2,) Observation: The input  resistance of a BJT/FET goes up in case we apply negative feedback using an emitter/source resitance.

I kindly ask you to answer two short questions:
a) Are both points 1) and 2) above, correct or not?
b) Can we derive from the observation in 2) any information about the question if a voltage or a current is fed back to the inverting node?

Are these questions "technically" enough?
I am awaiting your answer.

[T3sl4co1l]

If you talk amplifier circuits then yes you talk voltage because the feedback system uses voltage. but the transistor is a current controlled device. I have never heard it described otherwise. As soon as you stick a resistor in a BJT circuit you control the circuit (not the BJT) with voltage.

Simon please.  I know and understand the underlying physics.  I can't quote the equations verbatim anymore, but I can assure you, beta drops out as a side-effect of the voltage control.

If you still don't believe, try this circuit: logic level input, series diode, series current limiting resistor to base.  Emitter to ground, pullup to collector.  Measure turn-off time.

If it's current controlled, then as soon as base current ceases (without reversing -- the diode proves this), the collector current should cease.  But it doesn't, it takes quite a long time to change.

Whereas, if you drive it in a voltage mode (such as from a low impedance (but obviously not zero ohm) source, or another diode junction[1]), it goes quite quickly.  The usual approach being a simple B-E resistor to discharge base voltage on turn-off.

[1] As in the current mirror.  Which uses one diode-strapped BJT (note again, base current is tiny, collector current accounts for alpha times the total current) to drive the B-E of another transistor (which unless it's saturated, also draws little base current).

Tim

[IanB]

Further to what T3sl4co1l says, the currents through the device are controlled by charge accumulations in various regions inside it. An accumulation of charge is somewhat akin to a capacitance and the amount of charge varies with voltage. The base current is in a sense a parasitic current that occurs due to the way the device is built. You could wish it wasn't there, and in an FET that wish is granted.

[Zero999]

If you talk amplifier circuits then yes you talk voltage because the feedback system uses voltage. but the transistor is a current controlled device. I have never heard it described otherwise. As soon as you stick a resistor in a BJT circuit you control the circuit (not the BJT) with voltage.

Simon please.  I know and understand the underlying physics.  I can't quote the equations verbatim anymore, but I can assure you, beta drops out as a side-effect of the voltage control.

If you still don't believe, try this circuit: logic level input, series diode, series current limiting resistor to base.  Emitter to ground, pullup to collector.  Measure turn-off time.

If it's current controlled, then as soon as base current ceases (without reversing -- the diode proves this), the collector current should cease.  But it doesn't, it takes quite a long time to change.

Whereas, if you drive it in a voltage mode (such as from a low impedance (but obviously not zero ohm) source, or another diode junction[1]), it goes quite quickly.  The usual approach being a simple B-E resistor to discharge base voltage on turn-off.

That's the storage time. You'll only notice a difference, if the transistor is driven into saturation. If the series current limiting to the base is high enough, then the transistor will not go into saturation and will turn off quickly, even with the diode.

Another way to ensure it doesn't saturate is to with a Baker clamp. Add a diode with its cathode connected to the collector and the anode connected to the anode of the diode in series with the base. When the transistor starts to saturate, current is diverted away from the base, via the collector to the emitter, bypassing it.

[Simon]

So we are still back to driving a voltage through a resistor to create a current. Of course the hfe does change with different currents, as far as I know large collector currents (and therefore large base currents) cause the hfe to go down and I'm sure there will be other effects, that is the whole point of the feedback circuit and then people start claiming that the properties of the whole circuit are those of the BJT on it's own.

I rather get the impression that the illusive points and evidence that has not been properly explained amount to splitting hairs in the practical realm of electronics and we are not really here to look at physics and the original question was certainly not asking for it.

[c4757p]

So we are still back to driving a voltage through a resistor to create a current. Of course the hfe does change with different currents, as far as I know large collector currents (and therefore large base currents) cause the hfe to go down and I'm sure there will be other effects, that is the whole point of the feedback circuit and then people start claiming that the properties of the whole circuit are those of the BJT on it's own.

I get the distinct feeling that you have no idea what the rest of us are talking about.

[Simon]

Well maybe I'll find out one day 

[Tandy]

To be fair I have always understood thermionic valves as being voltage controlled and BJT to be current controlled subject to the required voltage to cross the junction. Clearly this received wisdom is not universally accepted and is causing a somewhat heated debate.

In all situations we are dealing with models that best represent the component operation and these models can never fully represent the physical properties of the component. Certainly during my education the current controlled model was taught so that is what I have used in my own work. It seems to have served me well over the years but I am always open to expanding my understanding where possible.

You will see that I have stayed quiet in this debate hoping to learn something from the debate but I feel it is rather adversarial and confusing. It would be interesting to know how these opposing opinions come about, is it due to limitations in the models used or a misunderstanding somewhere? I would appreciate more detailed explanations and links to references where possible because I am always happy to learn something new, even if that proves wrong something I have understood as being a fundamental principle all this time.

[Simon]

To be fair I have always understood thermionic valves as being voltage controlled and BJT to be current controlled subject to the required voltage to cross the junction. Clearly this received wisdom is not universally accepted and is causing a somewhat heated debate.

In all situations we are dealing with models that best represent the component operation and these models can never fully represent the physical properties of the component. Certainly during my education the current controlled model was taught so that is what I have used in my own work. It seems to have served me well over the years but I am always open to expanding my understanding where possible.

You will see that I have stayed quiet in this debate hoping to learn something from the debate but I feel it is rather adversarial and confusing. It would be interesting to know how these opposing opinions come about, is it due to limitations in the models used or a misunderstanding somewhere? I would appreciate more detailed explanations and links to references where possible because I am always happy to learn something new, even if that proves wrong something I have understood as being a fundamental principle all this time.

As I said i have never heard of this alternative explanation since I saw this thread.

[IanB]

To be fair I have always understood thermionic valves as being voltage controlled and BJT to be current controlled subject to the required voltage to cross the junction. Clearly this received wisdom is not universally accepted and is causing a somewhat heated debate.

In all situations we are dealing with models that best represent the component operation and these models can never fully represent the physical properties of the component. Certainly during my education the current controlled model was taught so that is what I have used in my own work. It seems to have served me well over the years but I am always open to expanding my understanding where possible.

You will see that I have stayed quiet in this debate hoping to learn something from the debate but I feel it is rather adversarial and confusing. It would be interesting to know how these opposing opinions come about, is it due to limitations in the models used or a misunderstanding somewhere? I would appreciate more detailed explanations and links to references where possible because I am always happy to learn something new, even if that proves wrong something I have understood as being a fundamental principle all this time.

There has been very little actual theoretical content in this thread (such content would have to contain an examination of physics, device models and equations). Here is a reference that goes over the theory in some detail: http://ecee.colorado.edu/~bart/book/book/chapter5/ch5_1.htm

In physics influences are usually conveyed and expressed in terms of fields. In the electrical/electronic domain we have electric fields and magnetic fields. Electric fields are produced by voltages and magnetic fields are produced by currents. Inside a transistor the various effects are mediated and regulated by electric fields produced by accumulations of charge in parts of the device, and thus by voltages. For a device to be controlled by current, we would have to find magnetic fields mediating the actions of that device. Magnetic fields are not significant in the operation of a transistor and therefore current in a transistor is not a controlling cause, but rather is a derived effect, or an outcome.

This is not to say that simplified models that assume base current to be a controlling cause are not useful, but such models are just that: models that make simplifying assumptions to suit certain purposes.

[dannyf]

is it due to limitations in the models used

Models don't create the underlying mechanism - which is the topic of the discussion. Instead, the underlying mechanism determines the models used.

If you look at the bjts, it is all about carriers (electronics and holes = negative charts + positive charges if you will) -> ie, it is a current - driven mechanism.

For the fets, it is about the electronic field that makes the majority carriers difficult to go through -> it is a voltage-driven mechanism.

All the discussion about tempco or diode equations or circuits are nonsensical and irrelevant for this discussion.

[free_electron]

Nobody knows at the deepest level how physics really works, all we have is various models of how materials behave.
What matters is which model is appropriate to our requirements.

Dave

not true. There exists equipment that can actually track electrons flowing in material (e-beam probing ) . IBM even has a machine that can arrange individual atoms to form the IBM logo... they have made a twenty atom transistor with it. and yes, it does work. they can track the electrons flowing. not simulated. measured.

e-beams are frequently used to sniff out leakage in integrated circuit design as you can count individual electrons.

[T3sl4co1l]

That's the storage time. You'll only notice a difference, if the transistor is driven into saturation. If the series current limiting to the base is high enough, then the transistor will not go into saturation and will turn off quickly, even with the diode.

Yes, storage time is a huge part of it, in saturated mode.  The collector can sit there drifting for tens of microseconds, even though the base terminal current is absolutely zero!

Another way to ensure it doesn't saturate is to with a Baker clamp. Add a diode with its cathode connected to the collector and the anode connected to the anode of the diode in series with the base. When the transistor starts to saturate, current is diverted away from the base, via the collector to the emitter, bypassing it.

This helps, but helps the most as a combination approach.

Just now I set up a circuit which I expect will meet your approval: the schematic as shown above, with 1k base resistor, 0/5V square wave input, a pair of 1N914, a 2N3904, and a 220 ohm collector load (to a +12V supply).

The waveform: well darn, my camera battery is low, so I'll have to describe it.  Anyway,

On the input falling edge, collector voltage does begin to rise immediately (the Baker clamp is indeed doing its job).  Starting from "saturation" (about 0.8V), for the first 270ns, it rises approximately linearly, to 4V.  Then the decay is approximately exponential, with a time constant of about 300ns.  The 10-90% rise is 680ns.

In the same time frame, base voltage barely falls at all, starting around 800mV and only falling to 600mV after a whopping 2.5us.  Given the Ebers-Moll equation says 60mV/decade, that means the collector current has only decreased by 2000x in that time, or reaching 25uA.  While that's quite a bit more "off" than it started, it still has a seriously long way to go before being truly cut off (~nA!).

Whereas if I simply add a B-E resistor of 1k, the input-falling propagation delay is 110ns, collector rise time is 85ns and base storage time is 160ns.

Still not convinced it's voltage not current?  Come on, you people are hopeless...

Tim

[T3sl4co1l]

Nobody knows at the deepest level how physics really works, all we have is various models of how materials behave.
What matters is which model is appropriate to our requirements.

Dave

not true. There exists equipment that can actually track electrons flowing in material (e-beam probing ) . IBM even has a machine that can arrange individual atoms to form the IBM logo... they have made a twenty atom transistor with it. and yes, it does work. they can track the electrons flowing. not simulated. measured.

e-beams are frequently used to sniff out leakage in integrated circuit design as you can count individual electrons.

Not only is electron science in general quite well understood, semiconductor physics itself has been quite well understood for much of the last century, with the advent of quantum mechanics.

In fact, the field-effect transistor was the first to be theoretically evaluated (back in the '30s I think), but it had to wait for materials science and process engineering to catch up (because of somewhat less well-understood theoretical aspects, which were nonetheless figured out later, which eventually factored into the development of practical commercial devices).

The BJT is fairly complex, theoretically speaking, but it was predicted and engineered in a quite reasonable manner.  It took until Bell Lab's famous model for a practical design to arrive, though triode type semiconducting devices were known before then (for instance, if you form two point-contact diodes very close together, you get majority charge carrier diffusion between the two junctions, and it acts as much more than simply two diodes on the same substrate!).

Tim