# EEVblog Electronics Community Forum

## Electronics => Beginners => Topic started by: raspberrypi on January 18, 2017, 01:14:59 am

Title: Current controlled device vs voltage controlled?
Post by: raspberrypi on January 18, 2017, 01:14:59 am
Current controlled device vs voltage controlled? What is the difference? I saw in MrCarlson's lab that he said a transistor was a current controlled device and a FET? was a voltage controlled device? I thought potential was potential. I know the difference in current and volts. But wouldn't this mean that a PNP was current controlled (Say it was used where you had the device then transistor then ground, so the pnp acts as a path to ground) and NPN was voltage controlled (you had the NPN then your load then ground so it feeding the load). You send voltage to the base of the NPN to turn on while current draw on the PNP turned it on.
Title: Re: Current controlled device vs voltage controlled?
Post by: danadak on January 18, 2017, 01:21:38 am
Bipolar, NPN or PNP, basic bipolar transistor operation is codfied,
low frequency, IC = beta x Ib, a simplified approximation, that is
base current controls current flow in the collector. Beta a characteristic
of the process, geometry, of the device. Varies a lot device to device.
And there are many other tertiary effects, but in simple form the above
suffices.

In the case of a JFET (saturation region) -

(https://wikimedia.org/api/rest_v1/media/math/render/svg/da74c7e5f1cd47fd5acb669ae5dcb25773c3c898)

Here Idrain controled by Vgs (device has an intrinsic pinch off voltage)

Regards, Dana.
Title: Re: Current controlled device vs voltage controlled?
Post by: TimFox on January 18, 2017, 01:48:49 am
In general, we distinguish devices according to the easiest description of the input.
For a resistor, current-driven and voltage-driven are equally easy.
For a vacuum tube or FET, the input impedance is very high so the input current is not well-determined.  They are best described as voltage controlled.
For a diode or BJT, the input voltage changes little as the input current changes a large amount, therefore they are best described as current controlled.
There is an entire field of non-linear component theory that defines non-linear resistors, etc. according to one variable (e.g. voltage) being a non-linear function of the control variable (e.g. current).
Title: Re: Current controlled device vs voltage controlled?
Post by: blueskull on January 18, 2017, 02:00:32 am
A MOSFET turns on "more" as gate to source voltage goes higher, there is virtually no current flowing into gate.
A BJT turns on "more" as base to emitter current goes higher, there is a virtually constant 0.6V across base and emitter.

If you want to know device physics, well, an N-channel MOSFET turns on by having more gate surface charge, the charge (voltage divided by capacitance) pulls holes (P dopant) from beneath P region, and injects electrons (N dopant), thus inverting the P region into a N region. Since source and drain are already N doped, if we change the channel to N doped as well, there will be only N doped regions, so reverse PN junction can't form, so current can flow freely.

For BJT, let's say an NPN transistor, if we inject current through base to emitter, the injected current is represented as a stream of electrons. To maintain equilibrium, a stream of holes will flow from emitter to base. Most of them will be too fast to be trapped by base region thus flowing into collector region, therefore we can control collector current by changing base current. In reality, for each carrier flowing from base to emitter, there will be more carriers generated from collector to emitter because each base to emitter carrier will knock off more carriers, thus providing current gain.
Title: Re: Current controlled device vs voltage controlled?
Post by: AG6QR on January 18, 2017, 02:22:24 am
Both the phrase "current controlled" and "voltage controlled" are simplifications.  But they are useful simplifications.

In the case of a current controlled device, the voltage is almost independent of the current, and the amount of current is what matters for control.

In the case of a voltage controlled device, the current is almost independent of the voltage, and the amount of voltage is what matters for control.

The phrase "almost independent" isn't the same as "completely independent".  But it's close enough to make it very difficult to reliably use a constant regulated voltage to control a current-controlled device, and likewise difficult to use a constant regulated current to reliably control a voltage-controlled device.
Title: Re: Current controlled device vs voltage controlled?
Post by: Brumby on January 18, 2017, 03:24:50 am
Both the phrase "current controlled" and "voltage controlled" are simplifications.  But they are useful simplifications.

In the case of a current controlled device, the voltage is almost independent of the current, and the amount of current is what matters for control.

In the case of a voltage controlled device, the current is almost independent of the voltage, and the amount of voltage is what matters for control.

The phrase "almost independent" isn't the same as "completely independent".  But it's close enough to make it very difficult to reliably use a constant regulated voltage to control a current-controlled device, and likewise difficult to use a constant regulated current to reliably control a voltage-controlled device.

I like this response as it answers the fundamental question.

I did wince a little bit with the phrase "almost independent" - but for the sake of getting the important points across, I would have to endorse it's use.

In regard to PNP or NPN bipolar transistors - both are current controlled.  It's just that the difference in polarity requires a little more thought in understanding things - especially if you've been working a lot with one type.  I can bias an NPN without blinking, but I have to stop and think if it's a PNP.
Title: Re: Current controlled device vs voltage controlled?
Post by: raspberrypi on February 13, 2017, 07:58:06 am
Both the phrase "current controlled" and "voltage controlled" are simplifications.  But they are useful simplifications.

In the case of a current controlled device, the voltage is almost independent of the current, and the amount of current is what matters for control.

In the case of a voltage controlled device, the current is almost independent of the voltage, and the amount of voltage is what matters for control.

The phrase "almost independent" isn't the same as "completely independent".  But it's close enough to make it very difficult to reliably use a constant regulated voltage to control a current-controlled device, and likewise difficult to use a constant regulated current to reliably control a voltage-controlled device.

I like this response as it answers the fundamental question.

I did wince a little bit with the phrase "almost independent" - but for the sake of getting the important points across, I would have to endorse it's use.

In regard to PNP or NPN bipolar transistors - both are current controlled.  It's just that the difference in polarity requires a little more thought in understanding things - especially if you've been working a lot with one type.  I can bias an NPN without blinking, but I have to stop and think if it's a PNP.

So what does that mean bias? I see where people have to "rebias" old radios when they put in silicon diodes from germanium. I always thought the NPN is more useful because it goes before the load; simplification:battery-> transistor ->load ->then ground, where as the PNP is the last step before ground so its less useful.
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 13, 2017, 09:16:01 am
Current controlled device vs voltage controlled? What is the difference? I saw in MrCarlson's lab that he said a transistor was a current controlled device and a FET? was a voltage controlled device?

For my opinion, this is a clear question - and there can be only one correct answer.
Physical laws do not accept different explanations.
Therefore, I cannot understand vague statements like:

"IC = beta x Ib, a simplified approximation, that is base current controls current flow in the collector. .... but in simple form the above suffices."

"For a diode or BJT, the input voltage changes little as the input current changes a large amount, therefore they are best described as current controlled."

"...therefore we can control collector current by changing base current"

"Both the phrase "current controlled" and "voltage controlled" are simplifications.  But they are useful simplifications."

No - we do not need any "simplifications" or "best descriptions".

The answer to raspberrypi`s question is: FET`s as well as BJT`s both are voltage-controlled!
I am aware that in some textbooks and many (too many!) forum contributions the BJT is described as current-controlled.
But this is simply WRONG!
This is just a correlation (and not a chain of causation) - which should better be read as IB=IC/B.
In words: IB is a (small) part of IC. Yes, there is a base current IB which - in most design cases - can/must be regarded as a finite value.
So what? Does this imply any control function?

There is only one fundamental control function - W. Shockley`s equation which describes the voltage-current relationships at a pn junction:
Ic=Is*exp[(VBE/VT)-1] .

All reliable knowledge sources (Universities of Berkeley and Stanford, MIT, ....) describe the BJT as voltage-controlled.
This applies also to high-quality books like "Art of Electronics" (Horowitz/Hill).
To me, this is the only logical description because a small current NEVER can directly control a larger current (even the well-known "water-flow" model cannot function this way from the energy point of view).

The working principle of many circuits (current mirror, differential amplifier...) and many observable effects (EARLY-effect, tempco of -2mV/K,...) can be explained only based on voltage-control.
Therefore, it is really a phenomenon to me that current-control is still under discussion. This view cannot be supported by any explanation - neither physical nor practical.

Two simple examples:
1) The gain expression of a common emitter stage contains the transconductance gm=IC/VT.   This parameter gm is nothing else than the SLOPE of the mentioned exponential function (Shockleys equation): gm=d(IC)/d(VBE). How can somebody deny that VBE controls IC?
2.) Question: Can somebody (who still is believing that IB would control IC) explain the function of a stabilizing resistor RE in the emitter path?

We do not need complicated physical considerations (behaviour of charged carrier etc.) - only simple observations of electrical circuits (and explanations of the observed effects) are sufficient  to understand that the BJT is, of course, a voltage-controlled device.

Title: Re: Current controlled device vs voltage controlled?
Post by: danadak on February 13, 2017, 11:58:40 am
A bipolar transistor can be controlled either as Vbe or Ib control.

Quote
The collector–emitter current can be viewed as being controlled by the base–emitter current (current control), or by the base–emitter voltage (voltage control). These views are related by the current–voltage relation of the base–emitter junction, which is just the usual exponential current–voltage curve of a p-n junction (diode).[1]

https://en.wikipedia.org/wiki/Bipolar_junction_transistor#Voltage.2C_current.2C_and_charge_control (https://en.wikipedia.org/wiki/Bipolar_junction_transistor#Voltage.2C_current.2C_and_charge_control)

"Normally" one designs to base control current. This is because of the non linear T dependent Vbe
behavior, not the least device to device variations issues.

The Ebers-Moll approximation for transistors -

This model shows how approximations can be used for simplified calculations, as well as more
exact.

Regards, Dana.
Title: Re: Current controlled device vs voltage controlled?
Post by: Sigmoid on February 13, 2017, 12:35:43 pm
Approximations for calculations is one thing, conceptual understanding is another...

I was just as much confused by "BJTs are current controlled" as the op of this thread.

The same with op amps, I've been royally confused over the things until I read Darren Ashby's book "Electrical Engineering 101" - where he breaks down the op amp into what it is: a difference operator and a huge gain in series. Now probably this only gave me my "eureka" moment because I have endured four semesters of being hit over my head with control loops, signals and linear systems, but it did for me what endless reiterations of "tries to keep the inputs at the same voltage", and similar rules of thumb didn't.

Of course I do use the "tries to keep the inputs at the same voltage" for calculating feedback loops. But the rule of thumb model for calculations and the definition that made me able to understand what the component does have been quite different.

So addressing BJTs and FETs, the way I understand it, "current controlled" just stands in for "has a low (and nonlinear) input impedance", and "voltage controlled" for "has an almost infinite input impedance". And since it has a low and nonlinear input impedance, it follows that you'll have to primarily manipulate current to get it to do exactly what you want it to do, while with an almost infininte input impedance, you can just put the input voltage on and you're set.
Title: Re: Current controlled device vs voltage controlled?
Post by: radiogeek381 on February 13, 2017, 12:38:12 pm
It is helpful to consider a BJT as voltage controlled when non-linear effects matter,
and current controlled when the conditions warrant.  (And charge controlled when
that's useful.)

Folks who do this for a living find all three models helpful at times.  Orthodoxy and
iconoclastic adherence may provide some comfort, but the inflexibility makes for
inefficient design.

Current controlled device vs voltage controlled? What is the difference? I saw in MrCarlson's lab that he said a transistor was a current controlled device and a FET? was a voltage controlled device?

For my opinion, this is a clear question - and there can be only one correct answer.
--- SNIP ---
The answer to raspberrypi`s question is: FET`s as well as BJT`s both are voltage-controlled!
I am aware that in some textbooks and many (too many!) forum contributions the BJT is described as current-controlled.
But this is simply WRONG!
This is just a correlation (and not a chain of causation) - which should better be read as IB=IC/B.

----- SNIP ----

There is only one fundamental control function - W. Shockley`s equation which describes the voltage-current relationships at a pn junction:
Ic=Is*exp[(VBE/VT)-1] .

----------------- SNIP ------------------

Two simple examples:
1) The gain expression of a common emitter stage contains the transconductance gm=IC/VT.   This parameter gm is nothing else than the SLOPE of the mentioned exponential function (Shockleys equation): gm=d(IC)/d(VBE). How can somebody deny that VBE controls IC?
2.) Question: Can somebody (who still is believing that IB would control IC) explain the function of a stabilizing resistor RE in the emitter path?

OK... In answer to the last question:
Consider a voltage source driving a resistor of Rb like this

GND -- Vs ---- Rb ---- B --- E --- Re --- GND

Now using the h-parameter model (that fixes the voltage between B and E
at a "nominal" diode drop), the voltage around the loop would be

Vs - (Rb*Ib + 0.7 + Re*Ic) = 0
If we take as our model a current controlled source Ic = hfe * Ib, then
Vs - 0.7 = Ib * (Rb + hfe * Re)

But for suitably large beta, any incremental change in Vs (call it "vs) causes an
additional current in Re: ie = vs/Re which is more-or-less equal to the
incremental collector current.  If the collector load is Rc, then the
incremental change in collector voltage will be Rc*vs/Re -- giving us
a gain of Rc/Re.

And I did it all with a current controlled model.  One could do the same
with a voltage controlled model, but you'd end up with a really unusable
analytic model.

----------------------------------

So that's question 2.

Now to the other part that looked interesting:

Quote
All reliable knowledge sources (Universities of Berkeley and Stanford, MIT, ....) describe the BJT as voltage-controlled.
This applies also to high-quality books like "Art of Electronics" (Horowitz/Hill).
To me, this is the only logical description because a small current NEVER can directly control a larger current (even the well-known "water-flow" model cannot function this way from the energy point of view).

The working principle of many circuits (current mirror, differential amplifier...) and many observable effects (EARLY-effect, tempco of -2mV/K,...) can be explained only based on voltage-control.
Therefore, it is really a phenomenon to me that current-control is still under discussion. This view cannot be supported by any explanation - neither physical nor practical.

1. All the reliable engineers I've ever worked with have used both models.
All the textbooks I've ever read employ several models.

2. The "the water flow model"?   I'd agree, nor does the "little elves" model, or the
"magic smoke" model.  I don't know anyone who uses either to describe the action of
a BJT.

However, please note that, as powerful as the voltage control model is, it does not
explain the behavior of an optically excited BJT.  Nor does it explain the triggering
of a parasitic BJT in CMOS circuits by injection of charges from nearby sources.
(I have seen both happen in real products.)  These are both charge controlled
phenomena.

Finally, the assertion that "a small current NEVER can directly control a larger current"
is offered as a support for the argument that "a small current does not control a larger current
in a BJT" which is a tautology -- a rhetorical trap for listeners.

In fact, a small current can cause the concentration of minority carriers in the base region
to become large enough to make it possible for current to flow in between the collector
and base, despite the presence of a reverse bias condition.  The number of carriers need
only be sufficient to reduce the barrier to further injection of carriers across the C-B barrier.
Once this is achieved, it is only necessary to maintain sufficient base current to support the
minority carrier concentration because of the few injected carriers that are swept into the
BE junction.

As long as we're appealing to the citadel of Berkely, see an EE105 lecture
https://inst.eecs.berkeley.edu/~ee105/sp04/handouts/lectures/Lecture22.pdf (https://inst.eecs.berkeley.edu/~ee105/sp04/handouts/lectures/Lecture22.pdf)
It rationally presents several models, each of which have some use in real
engineering and circuit design.

I don't particularly care what orthodoxy folks embrace, but I will point out that
adherence to one so rigid is neither efficient nor helpful for folks who want to
develop an understanding of how to design and explain the behavior of circuits.

respectfully,
Title: Re: Current controlled device vs voltage controlled?
Post by: RoGeorge on February 13, 2017, 12:41:07 pm
Causality in Electromagnetism strikes again!
:popcorn:

To be on-topic, even if this contradicts my own opinion that causality in electromagnetism is undefined, I'll say that
- bipolar transistors are current controlled
- field effect transistors are voltage controlled

Again, this voltage/current control classification I am voting for is just for practical and engineering reasons. Otherwise, a debate about the causality in electromagnetism will be interesting, but I'm not sure if causality makes any sense in the first place.
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 13, 2017, 12:43:52 pm
"Normally" one designs to base control current. This is because of the non linear T dependent Vbe
behavior, not the least device to device variations issues.

Can you, please, explain this statement in more detail? I rather think, that it is not VBE but the currents IE and IC which depend on the temperature.

Furthermore, I can confirm that the Ebers-Moll BJT-model (and the more advanced Gummel-Poon BJT-model) is based on voltage control.
More than that, they are the kernel of the BJT models used in simulation programs.
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 13, 2017, 01:10:43 pm
It is helpful to consider a BJT as voltage controlled when non-linear effects matter,
and current controlled when the conditions warrant.  (And charge controlled when
that's useful.)
.............
Folks who do this for a living find all three models helpful at times.

Well - it was not my intention to discuss several models which might exist.
And I did not want to discuss which view might be helpful or not.
Each student/beginner or designer can decide which approach he is willing to follow.

My only intention was to answer the main question: Is the collector current of the BJT determined by the voltage VBE or by the current IB.
It was my understanding that the question was related to the BJT as a part  - and not to a complete circuit where we sometimes use a large base resistor to realize something which we call "current injection".

And, in this context, only one single answer is possible - in spite of different existing small-signal models.

But for suitably large beta, any incremental change in Vs (call it "vs) causes an
I am sorry, but I cannot consider this as an answer to my question which was related to the stabilizing effects of RE (starting with an unwanted Ic increase due to tolerances or temperature effects).
More than that - where is the logic behind your statement that any change in the supply voltage Vs will cause an "additional current in Re" ?
Is Re directly connected to Vs? No - of course not. Why do you left out the most important part - the base-emitter path between Vs and Re?

1. All the reliable engineers I've ever worked with have used both models.
All the textbooks I've ever read employ several models.

Again - I do not speak about models. I speak about the BJT`s physical properties and the mechanism how IC can be varied.

Respectfully
LvW
Title: Re: Current controlled device vs voltage controlled?
Post by: danadak on February 13, 2017, 02:53:56 pm
Vbe is essentially a forward biased diode, and can be derived from Schockley's
equation (diode driven by a constant current).

Vd = n x Vt x ln( idiode / Isatdiode ). and Vt = kT/q

Regards, Dana.

Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 13, 2017, 03:55:23 pm
Vbe is essentially a forward biased diode, and can be derived from Schockley's
equation (diode driven by a constant current).
Vd = n x Vt x ln( idiode / Isatdiode ). and Vt = kT/q
Regards, Dana.
T
Basic law: No current without a driving voltage.
Physics: It is the VOLTAGE across a pn junction which influences the thickness of the depletion layer and, thus, allows a variable current.
We can manipulate each equation - but not change physical properties and dependencies.
Therefore: Shockley has introduced his formula only in this form (Vf=forward voltage): Forward current If=Io*[exp(Vf/Vt)-1]
Title: Re: Current controlled device vs voltage controlled?
Post by: RoGeorge on February 13, 2017, 04:13:52 pm
To simplify the debate, let's ask a simpler question:

We have an ideal resistor of a given value. The resistor is connected in a DC electric circuit. This will give us a voltage across the resistor, and also a current through the resistor. Again, all components are ideal, and all measured values are stationary values, and not zero.
For our given resistor, is the current controlled by the voltage, or is the voltage controlled by the current?
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 13, 2017, 04:53:32 pm
To simplify the debate, let's ask a simpler question:

We have an ideal resistor of a given value. The resistor is connected in a DC electric circuit. This will give us a voltage across the resistor, and also a current through the resistor. Again, all components are ideal, and all measured values are stationary values, and not zero.
For our given resistor, is the current controlled by the voltage, or is the voltage controlled by the current?

Only one correct answer is possible: The current is controlled by the voltage.
More than that - it is DETERMINED by the voltage.
I am aware that we all speak about something like a "voltage drop" caused by a current.
But this is a misconception - on the other hand, it helps to calculate and analyze electric circuits.
With other words: In electronics, there is nothing like a "current source" - we only can realize voltage sources having a large internal resistance.
Now - when we connect a load resistor to this "current source" we have, in fact, a driving voltage and a resistive VOLTAGE DIVIDER.
In each conductive body (like a resistor) we nead an E-field allowing moving of charges (which we call "current").
And such an E-field is the result of an applied voltage (in BOTH resistors).
A current never can "produce" a voltage. This would be a false understanding of Ohms law.

And this applies, of course, also to the B-E junction of a BJT. It is the voltage which determines the width of the depletion layer and, hence, the amount of emitter current IE which then is split into IB (small part) and IC (major portion).
Title: Re: Current controlled device vs voltage controlled?
Post by: RoGeorge on February 13, 2017, 08:13:01 pm
OK, but an electric current can happen as well without a voltage. Even more, we can also show how a current can "determine" the "appearance" of a voltage.

But how can we make the charges to flow without a voltage? It doesn't matter how, but if we manage to pump the charges in a circuit, then a voltage will be "produced".

It's not mandatory to have a voltage in order to move charges. We can pump charges in many ways, for example mechanically, as in a Van de Graaff generator. Charges are attached to the belt of the electrostatic machine. The belt is moving, so the charges are moving too. This will produce, by definition, a current. This current is as good as any other current, and if we push this current through our resistor, the voltmeter will indicate a voltage. In this situation, since all has started by the mechanical belt, I guess we can say the current is the cause of the voltage.

Another example of a current produced by something else than a voltage will be the "electrical wind" produced by a heated filament.
Another one will be the Seebeck effect.
Another one will be the photoelectric effect, and so on.

Almost any form of energy can be determined to "pump" some charges for us, and this flow of charges is exactly a current, which current will create a voltage. Now it seems like the current is the "cause" of everything, so there are no voltage controlled devices, just current controlled devices.

Can this be true? I guess not.
The logic fallacy here is, in my opinion, the assumption that one of them, either the current or the voltage, must determine or "produce" the other one.

Since both the voltage and the current are tied together inside the same equation, why not considering them as different faces of the same law?
Why there must be a causality between them?
And after all, how we define causality?
Title: Re: Current controlled device vs voltage controlled?
Post by: raspberrypi on February 13, 2017, 08:37:25 pm
The abbreviations are getting out of control. What do the abbreviations in those equations stand for? Biggest hurdle to learning is when you have to look up each abbreviation, then find you looked up the wrong one.  MrCarlson is never wrong!!!

Maybe I should just stick to organic chemistry, even though its technically a harder subject, for me its alot easier to understand then electronics.
Title: Re: Current controlled device vs voltage controlled?
Post by: rstofer on February 13, 2017, 08:41:35 pm
So what does that mean bias? I see where people have to "rebias" old radios when they put in silicon diodes from germanium. I always thought the NPN is more useful because it goes before the load; simplification:battery-> transistor ->load ->then ground, where as the PNP is the last step before ground so its less useful.

Biasing sets the operating point of the transistor.  I'm not going to paraphrase the much better explanations on Google but there are many including tutorials and videos.  Some bias conditions, like for switching, are trivial.  Class A amplifiers take a little more thought (as do the other amplifier configurations).
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 13, 2017, 09:36:41 pm
But how can we make the charges to flow without a voltage? It doesn't matter how, but if we manage to pump the charges in a circuit, then a voltage will be "produced".

OK - perhaps I should restrict my answer to the circuits and the parts we are speaking of in this thread: Simple elctronic circuits with BJT`s.
Let`s concentrate on our subject instead of discussing Van der Graaff generators and the Seebeck effect.

Our point was and still is: Which quantity controls/determines the collector current of a BJT, right?

It is not my intention to convince anybody - I just have tried to answer a question.
Of course, I do understand that nobody simply will believe me and my arguments because: Who am I already?

Therefore, I like to quote another - more reliable - person:

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

This quote comes from Barrie Gilbert - one of the worlds most known designer and inventor of novel BJT applications .
If somebody in the forum is interested to read more about the circuits and effects I have mentioned to supporti my view, I can communicate some details.
Title: Re: Current controlled device vs voltage controlled?
Post by: Zero999 on February 13, 2017, 11:40:25 pm
Sigh. Not this soporific argument again!  :horse:

Instead of debating about which is current or voltage controlled, how about helping the original poster how to understand transistors and the models used to describe them, rather than having a pissing contest?

And bear in mind this is the beginners section, so don't be too quick to post lots of maths.

Current controlled device vs voltage controlled? What is the difference? I saw in MrCarlson's lab that he said a transistor was a current controlled device and a FET? was a voltage controlled device? I thought potential was potential. I know the difference in current and volts. But wouldn't this mean that a PNP was current controlled (Say it was used where you had the device then transistor then ground, so the pnp acts as a path to ground) and NPN was voltage controlled (you had the NPN then your load then ground so it feeding the load). You send voltage to the base of the NPN to turn on while current draw on the PNP turned it on.
BJTs whether NPN or PNP behave in the same way. The difference is the polarities of the voltages and currents are reversed.

The same is true for P vs N channel FETs.

If you build the same circuit with the NPN transistors swapped with PNP transistors and vice versa, it will still work, as long as the polarity of all the power supplies and other polarised components such as diodes and electrolytic capacitors, is reversed.

FETs have a high impedance (open circuit) gate, consisting of a reverse biased diode (J-FET) or an insulating layer of polysilicon (MOSFET).

A BJT has a low impedance base, consisting of a diode junction.

Both BJTs and FETs need to be treated differently, depending on whether they're being used in the active region (analogue) or as a switch (digital).

In the case of both BJTs and FETs, when biased in the active region, changes in base-emitter (BJT) voltage or gate-source voltage, causes changes in the collector or drain current. In this mode, both of the transistors can be treated as voltage to current converters (transconductance  amplifiers).

When used as a switch, the transconductance is of less importance. To turn a BJT fully on, the base needs to be driven with a high enough current to support the required collector current. The base-emitter voltage is a parasite. It's generally assumed to be the highest value specified on the datasheet. In the case of a FET, the gate voltage needs to be high enough to achieve minimal drain resistance and once the transistor is on, no current is drawn from the gate.
Title: Re: Current controlled device vs voltage controlled?
Post by: retrolefty on February 14, 2017, 12:50:52 am
All I can state is that some post threads around here are just real timesinks. No real learning or teaching going on here, just ego exercises.   |O
Title: Re: Current controlled device vs voltage controlled?
Post by: rfeecs on February 14, 2017, 01:11:50 am
When used as a switch, the transconductance is of less importance. To turn a BJT fully on, the base needs to be driven with a high enough current to support the required collector current. The base-emitter voltage is a parasite. It's generally assumed to be the highest value specified on the datasheet. In the case of a FET, the gate voltage needs to be high enough to achieve minimal drain resistance and once the transistor is on, no current is drawn from the gate.

Good point that a lot of introductory articles and videos are about how to use transistor as a switch.  This is one source of "BJT is current controlled, FET is voltage controlled."

BTW, for MOSFETs the insulating layer is typically SiO2.  The gate is polysilicon.

Just to  :horse:, there are a lot of ways of looking at it, not just one true way.  Another way is "all transistors are charge controlled devices."  Yet another is "BJTs are current controlled" because the collector current is controlled by the emitter current.
Title: Re: Current controlled device vs voltage controlled?
Post by: RoGeorge on February 14, 2017, 03:05:56 am
Current controlled device vs voltage controlled? What is the difference?
Any voltage source can be seen as a current source, and any current source can be seen as a voltage source, so there is no real difference. They are equivalent. As long as you don't break the math, you can use whatever you like most, or whatever will ease your calculations.
See https://en.wikipedia.org/wiki/Source_transformation. (https://en.wikipedia.org/wiki/Source_transformation.) This is sometimes called the Thevenin-Norton equivalence.

I saw in MrCarlson's lab that he said a transistor was a current controlled device and a FET? was a voltage controlled device?
I don't know if you saw that, or not. I guess you saw. Anyway, FET stands for Field Effect Transistor, and BJT (which I guess will be the other kind of transistor from your question) stands for http://bfy.tw/A4bU (http://bfy.tw/A4bU)
Yes, people like to think about BJTs as current controlled devices (again, BJT means "normal" transistors, like NPN or PNP transistors) , and about FETs as voltage controlled devices. Because of Thevenin-Norton, it doesn't really matter how you want to see those transistors. Be careful here. Your teacher might want to see each device exactly as he, the teacher, teach you to see it. So, after all, no matter what Thevenin-Norton said, there is only one right answer: the one that your teacher is expecting from you.

I thought potential was potential. I know the difference in current and volts.
I have no idea what are you talking about.

But wouldn't this mean that a PNP was current controlled (Say it was used where you had the device then transistor then ground, so the pnp acts as a path to ground) and NPN was voltage controlled (you had the NPN then your load then ground so it feeding the load). You send voltage to the base of the NPN to turn on while current draw on the PNP turned it on.
No. For a device to be current controlled, it doesn't matter if the control current is going to the ground, or is coming from the load. It doesn't matter if you "push" a current into the Base of a NPN, or if you "pull" a current from the Base of a PNP. As long as you vary a current in order to control the transistor, then we are saying the transistor is current controlled.

In conclusion:
- What parameter are you tweaking when you want to control a PNP transistor?
- I am tweaking its Base current. Aha, a current! So this device is controlled by a current, so we call it a current controlled transistor.

- Now, what parameter are you tweaking when you want to control a NPN transistor?
- I am tweaking its Base current. Aha, a current! So this device is controlled by a current tweak, too, so a NPN transistor is also a current controlled transistor, like the PNP one.

- And finally, what are we tweaking when we want to control a FET transistor?
- Well, a Field Effect Transistor (AKA FET, or MOSFET) does not have a Base, in the first place. Instead of Base, the FET has a Gate. The Gate does not allow any current. Since the Gate has zero current, all we can tweak in order to control a FET is the voltage applied to its Gate. So, we say the FET is a voltage controlled transistor.

Title: Re: Current controlled device vs voltage controlled?
Post by: helius on February 14, 2017, 04:58:03 am
The gate of a FET is insulated from the channel, so current won't flow through it to the drain or source. But current still flows in and out of the gate because it has capacitance.
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 14, 2017, 08:52:41 am
Any voltage source can be seen as a current source, and any current source can be seen as a voltage source, so there is no real difference. They are equivalent. As long as you don't break the math, you can use whatever you like most, or whatever will ease your calculations.
.......
Yes, people like to think about BJTs as current controlled devices (again, BJT means "normal" transistors, like NPN or PNP transistors) , and about FETs as voltage controlled devices. Because of Thevenin-Norton, it doesn't really matter how you want to see those transistors.

I think, we shouldn`t mix circuit theorems and other mathematical manipulations/equivalences with real parts properties. The original poster was asking if the BJT is voltage or current controlled - that`s all.

- Now, what parameter are you tweaking when you want to control a NPN transistor?
- I am tweaking its Base current.
Sorry - but may I ask you: Did you ever design a BJT based amplifier stage?
What do you think about the following design steps

Design steps (common emitter stage with DC stabilization, Vcc given):
* Select IE (resp. IC) and collector/emitter resistors RC resp. RE,
* Calculate required base voltage VB - based on VE=IE*RE - and assuming a VOLTAGE VBE=(0.65...0.7) volts.
* Design voltage divider (rule of thumb: Divider current > 10*IB=10*IC/B).
* Voltage gain: G=-gm*RC/(1+gm*RE)
* Transconductance gm=d(IC)/d(VBE)=IC/VT
* Voltage gain does NOT depend on B (same DC quiescent point)

Question: Current control? Does everybody realize WHY we normally have select a divider current of of app. 10*IB ? (Answer: Because we want that the value of IB resp. the uncertainties/tolerances of IB play a minor role only and will cause just a small shifting of the desired DC operational point.)

Comment 1: IB was used only for providing the required voltage VB.
Comment 2: Unexpected IC increase reduces VBE (voltage feedback)

Fazit: Each designer and each newcomer/student should know WHY emitter degeneration (voltage feedback) is incoroporated - and HOW it works!

Attachement: The enclosed figures explain why it is sufficient to work with an estimate for VBE.

Title: Re: Current controlled device vs voltage controlled?
Post by: Zero999 on February 14, 2017, 09:08:13 am
BTW, for MOSFETs the insulating layer is typically SiO2.  The gate is polysilicon.
Yes, you're right. Polysilicon is a good conductor, which wouldn't be any good for an insulating gate. SiO2 is a good insulator.

Quote
Just to  :horse:, there are a lot of ways of looking at it, not just one true way.  Another way is "all transistors are charge controlled devices."  Yet another is "BJTs are current controlled" because the collector current is controlled by the emitter base current.
That's what you meant, wasn't it?

The original poster was asking if the BJT is voltage or current controlled - that`s all.
I don't think the question was that simple. There seemd to be some confusion in the first post.

"I know the difference in current and volts. But wouldn't this mean that a PNP was current controlled (Say it was used where you had the device then transistor then ground, so the pnp acts as a path to ground) and NPN was voltage controlled (you had the NPN then your load then ground so it feeding the load). You send voltage to the base of the NPN to turn on while current draw on the PNP turned it on. "

The abbreviations are getting out of control. What do the abbreviations in those equations stand for? Biggest hurdle to learning is when you have to look up each abbreviation, then find you looked up the wrong one.  MrCarlson is never wrong!!!

Maybe I should just stick to organic chemistry, even though its technically a harder subject, for me its alot easier to understand then electronics.

I struggled with organic chemistry, got an E at A-level. There are so many different reactions to remember. My biggest problem was I lost interest in the subject.

What transistor circuit are you trying to understand?

Title: Re: Current controlled device vs voltage controlled?
Post by: RoGeorge on February 14, 2017, 10:06:45 am
Sorry - but may I ask you: Did you ever design a BJT based amplifier stage?
Yes.

I understand your view, and I already considered it. I am not saying you are wrong.
In the same time, I would like to keep my interpretation, even if you don't agree with it. After all, who knows, maybe tomorrow I will realize I was wrong, or vice versa.

Until then, it's OK to disagree.
That is why I will not continue this debate.
Title: Re: Current controlled device vs voltage controlled?
Post by: Codebird on February 14, 2017, 10:17:15 am
Quote
No real learning or teaching going on here, just ego exercises.

Disagreements about how to teach. There are those who advocate the 'throw them in the deep end' approach, where the questioner is given a complete and detailed answer and expected to understand it. There are also the 'shallow end first' who will present simplified, easy-to-understand answers even if they are not entirely correct, because they are much easier to grasp and can then provide a foundation for learning the more accurate model later. Neither approach is really superior to the other.
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 14, 2017, 10:20:59 am
That is why I will not continue this debate.
Wise decision - same to me, because in a technical discussion I don`t like contributions like
........ rather than having a pissing contest?
Title: Re: Current controlled device vs voltage controlled?
Post by: Zero999 on February 14, 2017, 01:39:52 pm
Quote
No real learning or teaching going on here, just ego exercises.

Disagreements about how to teach. There are those who advocate the 'throw them in the deep end' approach, where the questioner is given a complete and detailed answer and expected to understand it. There are also the 'shallow end first' who will present simplified, easy-to-understand answers even if they are not entirely correct, because they are much easier to grasp and can then provide a foundation for learning the more accurate model later. Neither approach is really superior to the other.
The trouble is, in a situation like this, it just turns into a pissing contest and the person asking the question doesn't learn much. If people don't like others making statements like this, then they should accept others have different methods of teaching and people learn in different ways, rather than trying to convince others they're doing it wrong.

Personally I like a mixture of both methods. I appreciate being told that I'm being given a simplified explanation, rather than the full story, so I'm free to learn more in future.

As far the common emitter amplifier is concerned: I learnt the basics of that myself, before I went to college and I didn't worry about whether a BJT is voltage or current controlled. In fact I didn't even realise there was such a debate, until I started posting here.
Title: Re: Current controlled device vs voltage controlled?
Post by: MrAl on February 14, 2017, 02:57:49 pm
Hi,

This question can be answered in two basic ways:
1.  To the layman.
2.  To the theorist

To the layman, the BJT can be viewed as either current controlled or voltage controlled.  That's because in any given application one parameter may be MUCH more important than the other and the design gets a little easier if we do it one way rather than the other.

To the theorist, the voltage controlled BJT can apply but to be perfectly precise we have to limit that exact definition to the static case where we already have a voltage established.  That's because there is no way to establish a voltage without moving charge.  Also, i think (someone check this) that the current theory is that there is no question about what came first, the voltage or the current, as they both happen at the same time with no time lag between one or the other.  More current theory may have changed this view, but even if it did if the voltage comes first then the current can not be very far behind.  In practice it will never be measured by anyone who isnt in a laboratory trying to actually measure the difference.  Feel free to look this point up and see if you can verify or not.  But remember even if the voltage comes before the current it would have to be something like 0.001ps away from it or even less.

In many circuits however we also have the view of the BJT being "charge controlled', which brings in another view.  The Rutgers library has a book that goes into this in detail, and describes how to turn off a BJT really really fast by sweeping the charge carriers out of the base region.  What is that?  It's not explained on the basis of voltage it is explained on the basis of charge carriers because if you dont get them out of the base fast enough your transistor stays in saturation for too long.

My conclusion is because of all these views it is best to call a device "energy controlled" because except for the static case (which is not very interesting in practical circuits) we MUST move energy in order to change the characteristic of the device.

Still having said all that, the way we view the BJT mostly will depend on what design aspect we are targeting or what design formula we are using at the time.  We may end up using two views for the same project at some point.
Title: Re: Current controlled device vs voltage controlled?
Post by: Zero999 on February 14, 2017, 03:07:51 pm
That's because there is no way to establish a voltage without moving charge.  Also, i think (someone check this) that the current theory is that there is no question about what came first, the voltage or the current, as they both happen at the same time with no time lag between one or the other.
AC theory should answer that. It depends on the impedance. If it's inductive, the voltage comes first. If it's capacitive it's the current which is first.

In a transmission line current and voltage are anti-phase and which comes first, depends on where you take the measurement.
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 14, 2017, 05:17:54 pm
Disagreements about how to teach. There are those who advocate the 'throw them in the deep end' approach, where the questioner is given a complete and detailed answer and expected to understand it. There are also the 'shallow end first' who will present simplified, easy-to-understand answers even if they are not entirely correct, because they are much easier to grasp and can then provide a foundation for learning the more accurate model later. Neither approach is really superior to the other.

Hero999, with all respect - have you teaching experience? So - you really would give answers which are“not entirely correct“ - just because they are „simplified,  easy-to-understand answers“ ? Do you really think that students - who already have learned how a pn-diode works - are overstressed with the exponential function Ic=f(VBE) ?

If people don't like others making statements like this, then they should accept others have different methods of teaching and people learn in different ways, rather than trying to convince others they're doing it wrong. ...Personally I like a mixture of both methods. I appreciate being told that I'm being given a simplified explanation, rather than the full story, so I'm free to learn more in future.
Are we now discussing „teaching methods“ or do we try to find a suitable - and correct - answer to a pure technical question?

As far the common emitter amplifier is concerned: I learnt the basics of that myself, before I went to college and I didn't worry about whether a BJT is voltage or current controlled. In fact I didn't even realise there was such a debate, until I started posting here.

OK - many people do not know about the BJT`s working principle and they do not care. This is, of course, not a problem - however, when you were not aware of this „debate“ , how could you take part in this discussion and qualify my answers as „pissing contest“?
What do you think, could be the reason for the original question?
The questioner was confused because of different and contradictory explanations in different books (and, unfortunately, also in this thread).
I am afraid, he will be really disappointed because - as it seems - no pure technical knowledge exchange on a fair basis is possible.
Title: Re: Current controlled device vs voltage controlled?
Post by: rfeecs on February 14, 2017, 06:17:42 pm
Quote
Just to  :horse:, there are a lot of ways of looking at it, not just one true way.  Another way is "all transistors are charge controlled devices."  Yet another is "BJTs are current controlled" because the collector current is controlled by the emitter base current.
That's what you meant, wasn't it?

Actually, no.  I was referring to the argument that the way a BJT works is the collector current is controlled by the emitter current.  The base current is a side effect, the base-emitter voltage another side effect.  The amplification is due to the impedance transfer, turning a low impedance input into a high impedance output.  The "transfer-resistor" effect that the transistor is supposedly named after.
Title: Re: Current controlled device vs voltage controlled?
Post by: Zero999 on February 14, 2017, 07:22:44 pm
Quote
Just to  :horse:, there are a lot of ways of looking at it, not just one true way.  Another way is "all transistors are charge controlled devices."  Yet another is "BJTs are current controlled" because the collector current is controlled by the emitter base current.
That's what you meant, wasn't it?

Actually, no.  I was referring to the argument that the way a BJT works is the collector current is controlled by the emitter current.  The base current is a side effect, the base-emitter voltage another side effect.  The amplification is due to the impedance transfer, turning a low impedance input into a high impedance output.  The "transfer-resistor" effect that the transistor is supposedly named after.
Oh I see what you mean now. Thanks.

Disagreements about how to teach. There are those who advocate the 'throw them in the deep end' approach, where the questioner is given a complete and detailed answer and expected to understand it. There are also the 'shallow end first' who will present simplified, easy-to-understand answers even if they are not entirely correct, because they are much easier to grasp and can then provide a foundation for learning the more accurate model later. Neither approach is really superior to the other.

Hero999, with all respect - have you teaching experience? So - you really would give answers which are“not entirely correct“ - just because they are „simplified,  easy-to-understand answers“ ?
I have no professional teaching experiance but I have had to teach others electronics in the past. As far as providing not 100% accurate information is concerned: yes I do that. It's known as abstraction. For example, we're taught about Kerckhoffs current and voltage, laws, which we later find are incorrect when we go on to study transmission lines.

Quote
Do you really think that students - who already have learned how a pn-diode works - are overstressed with the exponential function Ic=f(VBE) ?
It depends on what level they've studied a diode. At the very beginning we were just taught that VF is a virtually constant 0.6V and changes very little with varying current. Again, this was an abstraction. Later on we studied the relationship between VF, IF and temperature.

Quote
If people don't like others making statements like this, then they should accept others have different methods of teaching and people learn in different ways, rather than trying to convince others they're doing it wrong. ...Personally I like a mixture of both methods. I appreciate being told that I'm being given a simplified explanation, rather than the full story, so I'm free to learn more in future.
Are we now discussing „teaching methods“ or do we try to find a suitable - and correct - answer to a pure technical question?
As far the common emitter amplifier is concerned: I learnt the basics of that myself, before I went to college and I didn't worry about whether a BJT is voltage or current controlled. In fact I didn't even realise there was such a debate, until I started posting here.

OK - many people do not know about the BJT`s working principle and they do not care. This is, of course, not a problem - however, when you were not aware of this „debate“ , how could you take part in this discussion and qualify my answers as „pissing contest“?
What do you think, could be the reason for the original question?
The questioner was confused because of different and contradictory explanations in different books (and, unfortunately, also in this thread).
I am afraid, he will be really disappointed because - as it seems - no pure technical knowledge exchange on a fair basis is possible.
We don't know why the original poster asked the question. Very little background information was given. I don't think it's necessary to explain to them how a transistor works in the minutest of detail, down to what's going on with the charge carries in the semiconductor junction, however technically accurate it may be. All it would do is create more confusion. They already said they're confused about the difference between a PNP and an NPN transistor, so perhaps we should straighten that out before going any further?

It's quite likely a lot of what has been posted here is completely irrelevant to the original poster's question and will just cause more confusion.
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 15, 2017, 08:59:57 am
We don't know why the original poster asked the question. Very little background information was given. I don't think it's necessary to explain to them how a transistor works in the minutest of detail, down to what's going on with the charge carries in the semiconductor junction, however technically accurate it may be.

Who has done it?
In which contribution the transistor was described "in the minutest of detail, down to what's going on with the charge carries in the semiconductor junction" ?

The situation is as follows:
There was a question from raspberrypi with the title: Current controlled device vs voltage controlled?

"..IC = beta x Ib, a simplified approximation, .... but in simple form the above suffices."

"...they are best described as current controlled."

"...but they are useful simplifications."

To me, an engineer resp. somebody who wants to become an engineer cannot be satisfied with such vague statements. The problem is even more complex because the questioner will notice that in different books different claims (current-controlled, voltage-controlled) can be found - often without explanations or justifications.

That was the background of my answer to the questioner (post#7).
In this post, I gave a definite answer - according to my best knowledge and my experiences.
My reply was supplemented by some examples which should support the content of my contribution.

As an engineer, I expect that forum members who disagree with the contents of my answer, are able to present some counter arguments or - at least - give some technical comments to the purely circuit-oriented  facts I have mentioned (gain, transconductance, RE-voltage feedback, tempco -2mV/K...).

However, no such thing...
That`s somewhat disappointing.

Title: Re: Current controlled device vs voltage controlled?
Post by: MrAl on February 15, 2017, 10:49:01 am
That's because there is no way to establish a voltage without moving charge.  Also, i think (someone check this) that the current theory is that there is no question about what came first, the voltage or the current, as they both happen at the same time with no time lag between one or the other.
AC theory should answer that. It depends on the impedance. If it's inductive, the voltage comes first. If it's capacitive it's the current which is first.

In a transmission line current and voltage are anti-phase and which comes first, depends on where you take the measurement.

Hello,

That's not the same argument though as to which comes first.  You are saying that either voltage leads current or current leads voltage, which as you can see right away leads to a contradiction, because we can then ask the question, "So whcih is it?".

This argument is about having nothing to start with, no energy, and then trying to force a change of some kind.  To do that we have to separate charge and at the very start of that process it is believed that the electric field appears at the same moment that the charge is moved.  In other words, to move charge you need an electric field (void of any other input that is) and to create an electric field you need to move charge, so they both must happen at the same time with no delay between their appearance or start of their movement.

Also, if we energize an inductor with a step change we get a RAMP of current, not zero current, because the only time we see zero current is with zero voltage.  As soon as we get any voltage we get some current, even if it is very small.  This means simplified circuit analysis does not help understand this issue because at t=0 we think of having say 1 volt applied with zero current, but it is really t=0+ where we see the true effect.
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 15, 2017, 11:31:49 am
Also, if we energize an inductor with a step change we get a RAMP of current, not zero current, because the only time we see zero current is with zero voltage.  As soon as we get any voltage we get some current, even if it is very small.  This means simplified circuit analysis does not help understand this issue because at t=0 we think of having say 1 volt applied with zero current, but it is really t=0+ where we see the true effect.

Yes - completely agreed. While discussing time behaviour we must strictly distinguish between transient response and steady-state behaviuour.
AC analyses cannot tell us anything about delay ("first" or "second").  A phase of -30 deg is equivalent to a phase of +330deg.
Title: Re: Current controlled device vs voltage controlled?
Post by: danadak on February 15, 2017, 12:10:51 pm
Approximations vs inexact science.

I like the discussion because those who use approximations, like Ic = beta x Ib
(linear region) are just as effective as designers as those that spend days (well for
me it would be days) solving Poisons equation and the various transport equations
with doping levels, also all approximations, in the transistor.

Even the basic notion does the existence of V depend on I or vice versa. V of course
is a result of charge differential, static, but the charge had to be moved in free space to
create it. Horse before or after the cart.

After all we currently do not have a clue where dark matter comes from, just that it seems
to constitute an appreciable amount of mass in the universe. What is its effect on transistor
action one could posit.

Lastly here is stated a oversimplified explanation that a bipolar can be viewed either as current
or voltage controlled. Seems reasonable.

http://www.ece.umd.edu/~neil/enee313/BJTnotes_05_6_10.pdf (http://www.ece.umd.edu/~neil/enee313/BJTnotes_05_6_10.pdf)

Thanks ladies and gents, wonderful discussion. I am re-learning some stuff long forgotten
in my hummingbird sized brain pan.

Regards, Dana.
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 15, 2017, 04:32:34 pm
Lastly here is stated a oversimplified explanation that a bipolar can be viewed either as current
or voltage controlled. Seems reasonable.
.............
Thanks ladies and gents, wonderful discussion. I am re-learning some stuff long forgotten
in my hummingbird sized brain pan.
Regards, Dana.

Dana - I think, it is a good news to hear from you that the whole discussion (although not always very aim-oriented) had a certain value for you.
May I suggest something to you? From the text you have provided - carefully read only the parts after the figure again and try to find an answer for yourself to the following two questions:
(1) What do they mean with the frequently used term "bias" (voltage or current ?) and
(2) The text gives a rough explanation how the transistor works and how the collector current is produced - what do you now think is the primary quantity which determines the amount of collector current (VBE or IB)?
Title: Re: Current controlled device vs voltage controlled?
Post by: MrAl on February 16, 2017, 09:42:55 am
Hello again,

I think i see now what the main problem in this issue is.  It seems to stem from the key word, "control", or "controlled".

When we think of controlling something we think of something that is changing and we want to force some sort of change such that we get the desired response.  For example, when we drive a car down a highway we want to maintain a certain speed even if we hit some hills and travel up hill or down hill.  We might want to maintain say 60mph and so we vary the way we push on the accelerator pedal in order to maintain that speed.  So we are varying something to obtain some desired effect and we have to do that because of external conditions that change that we dont have any control over.

This differs from a phrase like, "mode of operation".  Let me give a simple example how these might be different.
Say we have two coils energized by two currents, and the coils are facing each other with fields opposing.  We know that they will repel each other.  If we turn one around, we know they will be attracted to each other.  How do we know this?  It's because we understand that the principle mode of operation is due to the magnetic field, and notice here we dont have to mention the electric field because we are assuming static operation.  But even so, we can say that it is the magnetic field that is responsible for this repel or attraction and we dont mention the electric field.
The catch is that it is hard to call this situation, "controlled".  In other words, it is hard to say that the magnetic field is "controlling" the attraction or repulsion.  By the definition of the word control it implies some sort of action such as in the non static case.  So here we find it more appealing to just say that the magnetic field is 'responsible' for the result.

In this same way i find it hard to say that the voltage is 'controlling' the transistor, however i dont have a problem with saying that the voltage is the principle mode of operation.  That is, certain things would happen as long as we maintain the voltage at the exact same level throughout, and do not try to change the current.  The catch we could run into when trying to say that it is 'controlled' by the voltage is if something else changes that is NOT under our control, we loose that control and we must change the drive signal to make up for it, and that is really controlling the transistor, and that requires a change of energy just like the car when we hit a hill that goes up or down.
If we look at a constant current source made from a bipolar transistor we quickly see that there is feedback and that means the base current (and voltage) will be constantly changing in order to properly 'control' the collector current.

So in conclusion i can say that i dont like stating that the transistor is controlled by voltage, but i dont mind saying that the principle mode of operation is based on voltage (in the static sense).  This means we never have to ask the question, "how did the voltage get there".

Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 16, 2017, 11:50:47 am
MrAl - I can fully agree to the first two paragraphs of your last contribution.
However, I have some objections against the rest of the text. Let me explain:

Quote: In this same way I find it hard to say that the voltage is 'controlling' the transistor, however I dont have a problem with saying that the voltage is the principle mode of operation. That is, certain things would happen as long as we maintain the voltage at the exact same level throughout, and do not try to change the current.

I cannot agree. The question if we are able to „maintain the voltage at the exact same level“ has nothing to do with physical determined properties of a device.

Quote: The catch we could run into when trying to say that it is 'controlled' by the voltage is if something else changes that is NOT under our control, we loose that control and we must change the drive signal to make up for it, and that is really controlling the transistor,

I am afraid that the working principles of a device do not depend on any changes which might be not „under our control“. In this context, may I direct your attention to the pdf figures I have attached to my reply’27 ? I think, the figures (stabilization lines) give an answer.
For my opinion, it would be a good thing when each designer of a BJT amplifier stage would be able to know the function of each part he is using.

Therefore - again my question:
How can the primary reason for using an emitter resistor RE (stablizing IC against unwanted changes, uncertainties and tolerances) explained WITHOUT using the relation between IE (resp IC)  and the voltage VBE?

Interestingly, I did get no answer in this thread up to now..
(The same applies to all other technical facts - verified by measurements (!) - I have mentioned in this thread supporting voltage control).

Title: Re: Current controlled device vs voltage controlled?
Post by: MrAl on February 16, 2017, 02:44:29 pm
MrAl - I can fully agree to the first two paragraphs of your last contribution.
However, I have some objections against the rest of the text. Let me explain:

Quote: In this same way I find it hard to say that the voltage is 'controlling' the transistor, however I dont have a problem with saying that the voltage is the principle mode of operation. That is, certain things would happen as long as we maintain the voltage at the exact same level throughout, and do not try to change the current.

I cannot agree. The question if we are able to „maintain the voltage at the exact same level“ has nothing to do with physical determined properties of a device.

Quote: The catch we could run into when trying to say that it is 'controlled' by the voltage is if something else changes that is NOT under our control, we loose that control and we must change the drive signal to make up for it, and that is really controlling the transistor,

I am afraid that the working principles of a device do not depend on any changes which might be not „under our control“. In this context, may I direct your attention to the pdf figures I have attached to my reply’27 ? I think, the figures (stabilization lines) give an answer.
For my opinion, it would be a good thing when each designer of a BJT amplifier stage would be able to know the function of each part he is using.

Therefore - again my question:
How can the primary reason for using an emitter resistor RE (stablizing IC against unwanted changes, uncertainties and tolerances) explained WITHOUT using the relation between IE (resp IC)  and the voltage VBE?

Interestingly, I did get no answer in this thread up to now..
(The same applies to all other technical facts - verified by measurements (!) - I have mentioned in this thread supporting voltage control).

Hi again,

I will say what i have said all along for the practical case, and that is that the type of analysis depends on what we want to know.  In some cases it is easier to use a voltage based equation and sometimes it is easier to use a current based equation.
I could have asked an equally valid question which would depend on knowing the base emitter voltage more so than the current, and that is in the practical case of the 'normal' voltage reference diode.  We use voltage because we care more about the voltage than the current.  If we cared more about the current we might want to know what the base current is, and knowing the spread of Beta we can design something that is independent of Beta.  However, we also can not forget that although we dont care as much about the current, the current still assumes whatever level it needs to be for the given device and that current will change with the voltage change.

Second to the point of "The question of if we are able to maintain the voltage at the exact level..."...
I actually said what you are saying too, but i only had a difference with what we called it, "control" vs "mode of operation".
My question to you is if you look at the two coil case (or even just two magnets really except that seems like there's no way to control it anyway) where we have the two coils facing each other and lets limit to the case where they are repelled by each other.  We know it is the magnetic field that causes the repulsion, but do you think it is a good idea to call the effect of the magnetic field "controlling" something?  For example, two magnets hanging by two threads each, suspended and repelling each other, is the magnetic field "controlling" the distance of separation?

So i agree that in order to understand the characteristic behavior we may use a voltage oriented formula, i just dont think it is a good idea to call it "controlling" just like with the magnetic field and the two coils or two magnets.

You can look up the definition of 'control' and see what you think.  I was going to ask if you could find other instances where it makes sense to call it controlling (other than the transistor) but i am not sure if i would be willing to accept them unless they made sense.

For the MOSFET, it is called, "voltage controlled", and most of us agree to that even though we can NEVER change the output without adding or taking away something physical from the input, which as we all know must be in the form of a current.
It is also interesting that this is the sole thing that causes so much confusion to people just getting into the field of electronics.  When they go to use a MOSFET, they almost always think that they can drive it with a 1 megohm resistor even if it has to switch at 100kHz.  What causes that confusion?  It's the description of the MOSFET as being "voltage controlled" and having high input impedance (another misnomer).

I realize that it is sort of a convention of sorts though to call something voltage controlled.  But i would bet we can not find one practical circuit that does not vary the voltage at the gate or base and thus the current as well.

So what do you say about the two coil or magnet example and the magnetic field?  Are the two magnets "field controlled" or not?

BTW you have kept this interesting :-)

Title: Re: Current controlled device vs voltage controlled?
Post by: Cerebus on February 16, 2017, 04:17:42 pm

I think you'll find that "iconoclastic" means almost exactly the opposite of what you think it means.
Title: Re: Current controlled device vs voltage controlled?
Post by: Cerebus on February 16, 2017, 04:20:48 pm
The gate of a FET is insulated from the channel, so current won't flow through it to the drain or source. But current still flows in and out of the gate because it has capacitance.

Ahem, JFETs aren't insulated, only MOSFETS (or IGFETs if you prefer that term).
Title: Re: Current controlled device vs voltage controlled?
Post by: 2N3055 on February 16, 2017, 07:20:02 pm
This is simple misunderstanding.

"current controlled" and "voltage controlled" were never meant to be exact technical and physical explanation of exact physical phenomena inside transistor.

It is a simplification, from introductory texts, to draw attention to difference between bipolar and field effect devices. And it was mentioned only in regards of static operation.

My professor used to say that bipolar was called "current controlled" to help us remember that in a circuit we will have to account that at static operating point we will have a current going into the base (or out of ) as opposed to "voltage controlled" field effect devices that will not draw (or source) any mentionable current from the gate. And in making a circuit , we have to be mindful of that and account for it.

And that is all.  At AC all bets are off, so we have real models for that...

One "current controlled" device I know of is magnetic relay...

And current and voltage is chicken and egg question.. They are inextricable from each other..
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 16, 2017, 08:54:04 pm
This is simple misunderstanding.

"current controlled" and "voltage controlled" were never meant to be exact technical and physical explanation of exact physical phenomena inside transistor.

It is a simplification, from introductory texts, to draw attention to difference between bipolar and field effect devices. And it was mentioned only in regards of static operation.

My professor used to say that bipolar was called "current controlled" to help us remember that in a circuit we will have to account that at static operating point we will have a current going into the base (or out of ) as opposed to "voltage controlled" field effect devices that will not draw (or source) any mentionable current from the gate. And in making a circuit , we have to be mindful of that and account for it.

And that is all.  At AC all bets are off, so we have real models for that...

One "current controlled" device I know of is magnetic relay...

And current and voltage is chicken and egg question.. They are inextricable from each other..

2N3055 - I am afraid you are in error.
Are you aware that you did nothing than to repeat a claim of your professor - without any attempt to justif your "belief" (it is nothing else).
Title: Re: Current controlled device vs voltage controlled?
Post by: 2N3055 on February 16, 2017, 10:08:49 pm
This is simple misunderstanding.

"current controlled" and "voltage controlled" were never meant to be exact technical and physical explanation of exact physical phenomena inside transistor.

It is a simplification, from introductory texts, to draw attention to difference between bipolar and field effect devices. And it was mentioned only in regards of static operation.

My professor used to say that bipolar was called "current controlled" to help us remember that in a circuit we will have to account that at static operating point we will have a current going into the base (or out of ) as opposed to "voltage controlled" field effect devices that will not draw (or source) any mentionable current from the gate. And in making a circuit , we have to be mindful of that and account for it.

And that is all.  At AC all bets are off, so we have real models for that...

One "current controlled" device I know of is magnetic relay...

And current and voltage is chicken and egg question.. They are inextricable from each other..

2N3055 - I am afraid you are in error.
Are you aware that you did nothing than to repeat a claim of your professor - without any attempt to justif your "belief" (it is nothing else).

And you kind of repeat yourself..

And I used my professor's quote because I can't say it better.. I don't like reinventing wheel...

I think you failed to understand what I wanted to say.
My stance (not a belief) is that those two terms are wrong, imprecise and of "linguistic nature" and should not be taken as a scientific statement of any sort.

If anything I agree with you.

Current doesn't activate bipolar transistor, Vbe does, but current WILL flow into base , and on a MOSFET it won't flow in a gate (static operating point, disregards charging of capacitances).
That is why some call it "current" or "voltage" to make that distinction... It's not explanation how it works but what it does.. And as I said, we used it only in that context, and only in an introductory course.. And my stance is that it was useful for that purpose at the time.

Your stance (or belief) is that there is no red color, but only 660 nm wavelength electromagnetic radiation.. I agree and understand your point.. But I will order red roses for my wife, because it is an OK approximation for that purpose... And I do order 660nm LED, or a laser.  So tool for the job.  Details and precision when appropriate...

It seems to me that you are confusing electronics engineers with theoretical solid state physicists and research scientists..
Many of us only use transistors, not many of us here work for semiconductor manufacturers and actually design and make transistors and ICs from silicon wafers. In engineering, simplification is welcome friend where appropriate.. Of course, only when appropriate...

I would like to thank you very much for invigorating discussion... It is nice to see people thinking hard and having passion about electronics...

Best regards,

Sinisa

Title: Re: Current controlled device vs voltage controlled?
Post by: rfeecs on February 17, 2017, 12:49:51 am
This quote comes from Barrie Gilbert - one of the worlds most known designer and inventor of novel BJT applications .

A little bit OT, but Barrie Gilbert developed a number of "current mode" circuits.  (Good thing he didn't call them "current controlled").

Here is his paper "Current Mode, Voltage Mode, or Free Mode? A Few Sage Suggestions":
http://cas.ee.ic.ac.uk/people/dario/files/E416/gilbert-voltagemode-currentmode.pdf (http://cas.ee.ic.ac.uk/people/dario/files/E416/gilbert-voltagemode-currentmode.pdf)

Here is a great issue of IEEE SSCS dedicated to Barry Gilbert  with some interesting articles by him about his early life as an engineer, including the development of "current mode" circuits.  It also contains a full reprint of his paper describing his famous four quadrant multiplier, a prime example of a "current mode" circuit.  Interesting that almost every equation describing it contains relations between the transistor currents.:

This is despite the fact that the only true way to look at it is those currents are the result of applied voltages.  There are no current sources.  Everything is controlled by voltage.  This is a fundamental law of the universe.  :blah: :blah: :blah:
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 17, 2017, 07:54:48 am

It seems to me that you are confusing electronics engineers with theoretical solid state physicists and research scientists..
Many of us only use transistors, not many of us here work for semiconductor manufacturers and actually design and make transistors and ICs from silicon wafers. In engineering, simplification is welcome friend where appropriate.. Of course, only when appropriate...
.......
I would like to thank you very much for invigorating discussion... It is nice to see people thinking hard and having passion about electronics...
Sinisa

Hi Sinisa - thank you for the last sentence. You have used the word "discussion", which - to me - means: Questions and answers, OK?
Therefore, I like to ask the following two questions:
1.) In which post I have made "confusing" explanations with "theoretical solid state physicists and research scientists" ?

2.) Regarding "simplification": All formulas and design strategies contain simplifications - this cannot be avoided.
Now my question: Can you agree that there is a big differeence between (a) simplifications (neglecting some minor influences) and (b) two different explanations for the working principle of a device ?

Comment: This difference wouldn`t matter too much if  it would not lead to some observable contradictions. However - as I have mentioned several times (but nobody reacts upon it) - there are many observable/measurable effects and circuit properties which can be explained with voltage control only: Transconductance, gain formula, RE-feedback, Early-effect, tan-h curve for diff. amplifier,....

I have some experience in teaching electronic fundamentals - and it has happened rather often that students came to me with questions like:
* Different books contain different explanations for the BJT`s working principle - which one is correct?
* How can the effect of RE-feedback be explained based on the assumption of current control?
* The voltage gain of a transistor stage does not depend on B resp. beta (same dc quiescent point) - how can the BJT be current-controlled?

That`s the reason I don`t think, the whole subject is a matter of "simplification" only - do you understand my point?

Title: Re: Current controlled device vs voltage controlled?
Post by: MrAl on February 17, 2017, 08:28:53 am
This quote comes from Barrie Gilbert - one of the worlds most known designer and inventor of novel BJT applications .

A little bit OT, but Barrie Gilbert developed a number of "current mode" circuits.  (Good thing he didn't call them "current controlled").

Here is his paper "Current Mode, Voltage Mode, or Free Mode? A Few Sage Suggestions":
http://cas.ee.ic.ac.uk/people/dario/files/E416/gilbert-voltagemode-currentmode.pdf (http://cas.ee.ic.ac.uk/people/dario/files/E416/gilbert-voltagemode-currentmode.pdf)

Here is a great issue of IEEE SSCS dedicated to Barry Gilbert  with some interesting articles by him about his early life as an engineer, including the development of "current mode" circuits.  It also contains a full reprint of his paper describing his famous four quadrant multiplier, a prime example of a "current mode" circuit.  Interesting that almost every equation describing it contains relations between the transistor currents.:

This is despite the fact that the only true way to look at it is those currents are the result of applied voltages.  There are no current sources.  Everything is controlled by voltage.  This is a fundamental law of the universe.  :blah: :blah: :blah:

Hi,

An interesting read or two.  I like the use of the word "sage" which according to the context of the paper would mean: "wise, especially as a result of great experience".

He starts from the perspective of *energy* also (sound familiar?), and digresses into current and/or voltage due to the fact that although energy is the prime mover we often simplify by assuming one or the other as a matter of convenience.
He also notes that current and voltage are translated through what we call resistance.
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 17, 2017, 08:29:22 am
Hi MrAl,

I will say what i have said all along for the practical case, and that is that the type of analysis depends on what we want to know. In some cases it is easier to use a voltage based equation and sometimes it is easier to use a current based equation.

Here, the "case" is: How can we explain the stabilization effect of RE?.
Why didn`t you give a direct answer to this technical question?
Up to now - nobody from the "current-control party" has answered this simple question (let alone other questions regarding transconductance, etc).

I realize that it is sort of a convention of sorts though to call something voltage controlled.  But i would bet we can not find one practical circuit that does not vary the voltage at the gate or base and thus the current as well.
I think, nobody (including me) has denied the existence of a base current. But - does this automatically means that this current does physically controls (determines) the collector current? Remember how the simple relation Ic=B*IB came into being:
Because of IC=alpha*IE we have IB=(1-alpha)*IE and IB=[(1-a)/a]IC=(1/beta)IC.
Physically, this means: IB is a small part of IC.  Of course, we can rewrite the relation as IC=beta*IB - but this does not automatically allow us to interpret this as a ocntolling function (in the sense of "cause and effect").

So what do you say about the two coil or magnet example and the magnetic field?  Are the two magnets "field controlled" or not?

I am sorry, I cannot say anything to this example. I am not familiar with this subject - and, at the moment, I have not enough time to think about it (with the aim to give a substantial answer).
Regards
LvW
Title: Re: Current controlled device vs voltage controlled?
Post by: MrAl on February 17, 2017, 08:41:33 am
Hi MrAl,

I will say what i have said all along for the practical case, and that is that the type of analysis depends on what we want to know. In some cases it is easier to use a voltage based equation and sometimes it is easier to use a current based equation.

Here, the "case" is: How can we explain the stabilization effect of RE?.
Why didn`t you give a direct answer to this technical question?
Up to now - nobody from the "current-control party" has answered this simple question (let alone other questions regarding transconductance, etc).

I realize that it is sort of a convention of sorts though to call something voltage controlled.  But i would bet we can not find one practical circuit that does not vary the voltage at the gate or base and thus the current as well.
I think, nobody (including me) has denied the existence of a base current. But - does this automatically means that this current does physically controls (determines) the collector current? Remember how the simple relation Ic=B*IB came into being:
Because of IC=alpha*IE we have IB=(1-alpha)*IE and IB=[(1-a)/a]IC=(1/beta)IC.
Physically, this means: IB is a small part of IC.  Of course, we can rewrite the relation as IC=beta*IB - but this does not automatically allow us to interpret this as a ocntolling function (in the sense of "cause and effect").

So what do you say about the two coil or magnet example and the magnetic field?  Are the two magnets "field controlled" or not?

I am sorry, I cannot say anything to this example. I am not familiar with this subject - and, at the moment, I have not enough time to think about it (with the aim to give a substantial answer).
Regards
LvW

Hi again,

(Bipolar RE) What kind of answer are you looking for here, i would be happy to reply again :-)
But real quick, cant that be answered on the basis of the voltage difference across the base resistor (keeping the circuit simple) and that happens even if we keep the Vbe completely *constant* with no change in that itself.  That would immediately tell us that Vbe is not doing much if anything at all.  I will rectify this though if you want a different kind of answer.

Also, I posted one more time before your most recent reply, where i noted that in that Gilbert paper he starts from the perspective of *energy*.  Now why would anyone want to go and do something like that :-)

I am happy you are sticking with this as your interest is appreciated and your view is of course interesting as well.

Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 17, 2017, 08:47:55 am
A little bit OT, but Barrie Gilbert developed a number of "current mode" circuits.  (Good thing he didn't call them "current controlled").

I know these papers - and some others dealing with the same subject.
However - I think, we clearly must distinguish between "circuits" and a single part like a "transistor".
Of course, there are complete circuits which we operate in a mode called "current mode" (Current conveyor, current-feedback amplifier,..) .
However, we are describing these circuits - more or less - from a practical point of view  (independent on transistor physics).
Summary: All of these papers do not provide any arguments for or against the working principle of a single transistor.

What is, in this context, the meaning of "practical"?
Simple example: Even all "defenders" of BJT`s current control consider an opamp (with a BJT input stage) as a voltage-controlled circuit. Why? Because - from the practical point of view - it makes sense (and is allowed in most cases) to neglect the input current (although they believe that this current is the controlling quantity). Do you understand what I mean?

Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 17, 2017, 08:59:22 am
(Bipolar RE) What kind of answer are you looking for here, i would be happy to reply again :-)

MrAl, I am sure, you will be not surprised to read my answer:
Any unwanted IC increase (temperature, tolerances) above the design value will increase the emitter voltage VE=IE*RE and, thus, reduce the base-emitter voltage VBE. This assumes a relatively constant ("stiff") base voltage VB. And exactly THIS requirement is the reason we design a voltage divider (for providing VB) with resistors which are as low as possible (rule of thumb: Divider current 10*IB) - with respect to other design constraints (power consumption, circuits input resistance). Where is the role of the (controlling?) base current?
I think, these are the commonly accepted and applied design rules - am I wrong?
In this context, I like to remind you again to my attachements in post#27. Both figures show and explain what`s really happening and how the bias points are determined.

(One final comment. You wrote: "I am happy you are sticking with this as your interest is appreciated and your view is of course interesting as well."
I like to mention that this is not only "my view" - this view is shared by high-quality books (e.g. Art of Electronics) and leading US-universities.)
Title: Re: Current controlled device vs voltage controlled?
Post by: MrAl on February 17, 2017, 01:45:15 pm
(Bipolar RE) What kind of answer are you looking for here, i would be happy to reply again :-)

MrAl, I am sure, you will be not surprised to read my answer:
Any unwanted IC increase (temperature, tolerances) above the design value will increase the emitter voltage VE=IE*RE and, thus, reduce the base-emitter voltage VBE. This assumes a relatively constant ("stiff") base voltage VB. And exactly THIS requirement is the reason we design a voltage divider (for providing VB) with resistors which are as low as possible (rule of thumb: Divider current 10*IB) - with respect to other design constraints (power consumption, circuits input resistance). Where is the role of the (controlling?) base current?
I think, these are the commonly accepted and applied design rules - am I wrong?
In this context, I like to remind you again to my attachements in post#27. Both figures show and explain what`s really happening and how the bias points are determined.

(One final comment. You wrote: "I am happy you are sticking with this as your interest is appreciated and your view is of course interesting as well."
I like to mention that this is not only "my view" - this view is shared by high-quality books (e.g. Art of Electronics) and leading US-universities.)

Hi,

Am i wrong or does it seem like you wont accept any current based theory on the bipolar transistor?
It seems like you are just against any analysis that uses current as a basis for the calculations, such as feedback and bias.
I am probably wrong here, but it would be good to hear YOUR version of a current based theory on say how to bias the transistor.  Did you actually ever use one based on current?
I ask because i would like to hear your view on when we can use current based models/calculations.

As to the references you posted, i will check those out ASAP maybe i missed something.
But i can tell you that sometimes theory is not quite stated correctly or rather i should say 'completely'.  For example, in theory sometimes a short is not a zero Ohms resistance, sometimes it can be both a zero Ohms resistance and an open circuit at the same time.  This is interesting too because for current we consider this kind of short a zero Ohm resistance but for voltage we consider it an open circuit.  In the wording of the theory however it will not say anything except "short".
This brings up another question though, and that is what did you think about Gilbert's idea of "current mode" and "voltage mode", why do you think he called them that?

LATER:
In #27 it looks like you are saying that the change in Vbe is small?  Perhaps you can clarify.
I was going to suggest a mechanical model that can be easily understood which is like an analogy to the BJT.
If we have a large lever with a fulcrum with a sharp edge, one end of the lever goes down when we press down on it and the other end goes up.  If the end we press on has length 2 meters from the fulcrum and the other end has length 20 meters, when we press down 1cm on the 2 meter end the 20 meter end goes up by 10cm.  It's a simple gain in distance.
Now we change the fulcrum to one that has a logarithmic or exponentially shaped top surface rather than a sharp edge and it relatively short (say 1cm which is valid).  When we press on the 2 meter end now we get not only the gain of 10 as before, but the strange top causes an extra lift at the center thereby raising the end a little more.  But more to the point, the lever beam tracks the fulcrum strange surface as it tilts, and thus we can map out the behavior based on the position of where the lever beam TOUCHES the new fulcrum top surface, and there will be a one to one correspondence of that position to the height of the 20 meter end.
Now in the Vbe view of the transistor as compared to this lever, we would be looking at the minute change in position along the short topped fulcrum in order to calculate the height of the 20 meter end.  Does that make sense?  I dont think so.  It  is completely valid, in theory, to do so, but it just doesnt make sense.  It is better to track the 2 meter end and thereby compute the height of the 20 meter end.
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 17, 2017, 03:15:36 pm
Am i wrong or does it seem like you wont accept any current based theory on the bipolar transistor?
It seems like you are just against any analysis that uses current as a basis for the calculations, such as feedback and bias.
I am probably wrong here, but it would be good to hear YOUR version of a current based theory on say how to bias the transistor.  Did you actually ever use one based on current?
I ask because i would like to hear your view on when we can use current based models/calculations.

The answer is very simple (and I gave it already in my post#27):
(For your convenience, here I repeat my steps for designinmg a simple emitter stage using the currents IC, IE and IB)

Design steps (common emitter stage with DC stabilization, Vcc given):
* Select IE (resp. IC) and collector/emitter resistors RC resp. RE,
* Calculate required base voltage VB - based on VE=IE*RE - and assuming a VOLTAGE VBE=(0.65...0.7) volts.
* Design voltage divider (rule of thumb: Divider current > 10*IB=10*IC/B).
* Voltage gain: G=-gm*RC/(1+gm*RE)
* Transconductance gm=d(IC)/d(VBE)=IC/VT (using the exponenetial equation IC=f(VBE)
* Voltage gain does NOT depend on B (same DC quiescent point)

Question: Current control? Does everybody realize WHY we normally have select a divider current of of app. 10*IB ? (Answer: Because we want that the value of IB resp. the uncertainties/tolerances of IB play a minor role only and will cause just a small shifting of the desired DC operational point.)

Comment 1: IB was used only for providing the required voltage VB.
Comment 2: Unexpected IC increase VE=IE*RE and reduces VBE (current controlled voltage feedback)

But i can tell you that sometimes theory is not quite stated correctly or rather i should say 'completely'.
This brings up another question though, and that is what did you think about Gilbert's idea of "current mode" and "voltage mode", why do you think he called them that?
If you read my contributions again, you will notice that I did NOT need theoretical considerations for explaining my arguments - in contrary, observable and measurable effects only. (Again: Up to now, nobody gave any comments to these examples!)

I think, it was not B. Gilberts "idea" to create the terms "current mode". It was already invented and used earlier (in conjunction with the speed of analog signal processing). I do not intend to discuss these terms here, because there is no direct connection to the subject here under discussion. I have explained this before (these terms are primary related to signal processing properties within a complex circuit).

LATER:
In #27 it looks like you are saying that the change in Vbe is small?  Perhaps you can clarify.
No - this was not my saying. In contrary, everybody can see what happens for good stabilization: When the slope of the stabilization line is small (and at the vertical axis one can see how to make it small), the IC uncertainty d(IC) is also small - in spite of a relatively large (allowed) VBE uncertainty d(VBE). Of course, this is not a new finding but can be found in many books.
But it is an important property of such circuits because very often I`ve heard the question (not only from beginners):
How can VBE determine the current IC when it is (a) not known exactly and (b) assumed to be a constant value ??
Title: Re: Current controlled device vs voltage controlled?
Post by: MrAl on February 17, 2017, 06:34:34 pm
Hi,

Actually i meant did you ever do something like:
Ic=Ib*Beta

and be happy with it without invoking Vbe or any voltage for that matter?
Title: Re: Current controlled device vs voltage controlled?
Post by: rfeecs on February 17, 2017, 07:19:20 pm
With regard to language and definitions, just for fun I went back and looked at the textbook that UC Berkeley used when I went there way back when.  It was "Basic Circuit Theory" by Desour and Kuh, 1969 edition.

Attached is an excerpt where they talk about non-linear resistors and define the terms "current controlled" and "voltage controlled".  They define a "voltage controlled" resistor as one where the current is a single valued function of the voltage.  A "current controlled" resistor is one where the voltage is a single valued function of the current.  So a linear resistor is both "voltage controlled" and "current controlled".  The same applies to an ideal diode.  A tunnel diode would be "voltage controlled" but not "current controlled".

Clearly their definition is not the same as LvW's or Mr. Carlson's.

Another consideration is circuit theory classes dealt strictly with mathematical models of circuit elements.  They did not go into the physics of the devices.  The physical operation of circuit elements was covered in physics classes.

So apart from clearly defining your terms, it is also important to distinguish whether you are talking about the physics of device operation vs just the purely mathematical model of the device.
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 17, 2017, 09:24:52 pm
Hi,
Actually i meant did you ever do something like:
Ic=Ib*Beta
and be happy with it without invoking Vbe or any voltage for that matter?

Ohh - such a short reply (in fact, not an answer) to my last contribution.
Let me think about your question.......no, as far as I remember: Only in the form Ib=Ic/beta because I always select at first a suitable collector current and treat the base current Ib as a result. No doubt about it - the base current does exist. However, for some designs it can even be neglected (I have already mentioned the opamp).
May ask you back - in which situation did YOU use the mentioned relation in the form given in your question (without invoking the voltage Vbe) ??
Title: Re: Current controlled device vs voltage controlled?
Post by: rfeecs on February 17, 2017, 10:18:16 pm
Another fun thing. :)

Here is a video of Barrie Gilbert discussing translinear circuits where he says, around time 7:45, to forget what you have learned about a transistor as a current controlled current source... the best way to view the transistor at least for the purposes of this discussion is as a voltage controlled current source.

https://youtu.be/LQNJVtcFrCc (https://youtu.be/LQNJVtcFrCc)
Title: Re: Current controlled device vs voltage controlled?
Post by: MrAl on February 18, 2017, 07:16:10 am
Hi,
Actually i meant did you ever do something like:
Ic=Ib*Beta
and be happy with it without invoking Vbe or any voltage for that matter?

Ohh - such a short reply (in fact, not an answer) to my last contribution.
Let me think about your question.......no, as far as I remember: Only in the form Ib=Ic/beta because I always select at first a suitable collector current and treat the base current Ib as a result. No doubt about it - the base current does exist. However, for some designs it can even be neglected (I have already mentioned the opamp).
May ask you back - in which situation did YOU use the mentioned relation in the form given in your question (without invoking the voltage Vbe) ??

Hello again,

Yeah, when i ask someone a simple generic question like if they ever used:
"c=a*b"

and they feel compelled to answer:
"No, i only used it as a=c/b"

I know the conversation is not going anywhere.

There are some diehard advocates of one theory over another, and that's entirely up to them.  It is always up to the individual what they want to use and i wont try and take that away from them.

I used the form:
Beta=iC/iB

many times to test a bipolar for basic operation, if that makes you happy :-)
The Beta can change drastically in some transistors if they have been abused.  If the Beta is too low you know something is wrong right away.  And no, i did not have to measure Vbe and see if the reading was 0.5125 instead of 0.5124 i only had to see that when i APPLY 1ma to the base i get at least 20ma or higher in the collector.

But as i said before, go ahead and feel free to measure the Vbe voltage, it's always your choice.  Maybe you can get all the transistor companies to stop publishing the Beta values on the data sheets too.

Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 18, 2017, 08:44:40 am
MrAl, with all respect - what is the purpose of such polemic words like
Maybe you can get all the transistor companies to stop publishing the Beta values on the data sheets too.“

Why do you completely ignore my use of  B resp. beta values in all of my posts ? Didn`t you notice the appearance of  B in the design steps listed in my post#59 and in my drawings (post#27) ?
Do you recognize the B value on the vertical axis of the 2nd figure? This B value determines the slope of the stabilization line! The same applies to the 1st figure because the shown factor alpha is the basis for defining beta. Didn`t you recognize this?

Quote: „But as i said before, go ahead and feel free to measure the Vbe voltage, it's always your choice.“

May I ask you: Why do you insinuate that I would suggest to measure Vbe?  Didn`t you understand the meaning and the background of my drawings in post#27 ? Both figures are nothing else than a visual representation of well-known formulas for designing amplifier stages.
OK - I will again explain it to you: Both figures demonstrate why it is sufficient to assume a Vbe guess value of app. 0.7V - WITHOUT knowing the exact values !

Sorry to say - but I very much regret that it seems not to be possible to discuss on a fair and factual basis using exclusively pure technical questions and answers (answers and comments - for instance  - to the technical examples I have given in my various posts).
LvW

PS: I have discussed this subject (current vs, voltage controlled) not for the first time.
And each time I am really surprised that this subject causes a very emotionally charged controversy.
A real exchange of technical arguments appears not to be possible.
(Perhaps, because the "current-control believer" have nothing else than one single equation Ic=B*IB that - in my opinion - is misinterpreted as a false sequence of cause and effect ? Whereas voltage control is supported by many observations and measurable circuit properties.)

Title: Re: Current controlled device vs voltage controlled?
Post by: MrAl on February 18, 2017, 09:08:33 am
MrAl, with all respect - what is the purpose of such polemic words like
Maybe you can get all the transistor companies to stop publishing the Beta values on the data sheets too.“

Why do you completely ignore my use of  B resp. beta values in all of my posts ? Didn`t you notice the appearance of  B in the design steps listed in my post#59 and in my drawings (post#27) ?
Do you recognize the B value on the vertical axis of the 2nd figure? This B value determines the slope of the stabilization line! The same applies to the 1st figure because the shown factor alpha is the basis for defining beta. Didn`t you recognize this?

Quote: „But as i said before, go ahead and feel free to measure the Vbe voltage, it's always your choice.“

May I ask you: Why do you insinuate that I would suggest to measure Vbe?  Didn`t you understand the meaning and the background of my drawings in post#27 ? Both figures are nothing else than a visual representation of well-known formulas for designing amplifier stages.
OK - I will again explain it to you: Both figures demonstrate why it is sufficient to assume a Vbe guess value of app. 0.7V - WITHOUT knowing the exact values !

Sorry to say - but I very much regret that it seems not to be possible to discuss on a fair and factual basis using exclusively pure technical questions and answers (answers and comments - for instance  - to the technical examples I have given in my various posts).
LvW

PS: I have discussed this subject (current vs, voltage controlled) not for the first time.
And each time I am really surprised that this subject causes a very emotionally charged controversy.
A real exchange of technical arguments appears not to be possible (perhaps, because the "current-control believer" have nothing else than one single equation Ic=B*IB that - in my opinion - is misinterpreted as a false sequence of cause and effect ? Whereas voltage control is supported by many observations and measurable circuit properties.)

Hi,

I have to admire your patience.  I may have lost mine just a little bit here while you have kept yours intact so i must admire that.

I am afraid that i just dont understand your BASIC line of reasoning.  You're saying on the one hand that you believe in BJT voltage controlled, yet you are keeping Vbe constant.  There are some other things too but i am trying to keep each reply as simple as possible.  I would even, if i could, keep it down to a 'yes' or 'no' answer so that we could get to the bottom of this :-)

For example, and i am keeping this short and simple, do you believe that the form Ic=Beta*Ib is useful?
I think your answer is 'yes' from what you said, and you also said that Vbe doesnt matter.  If Vbe does not matter than why try to push voltage control?

Please try to keep the reply down as short as possible as this makes it easier to understand and reply.  Thanks in advance.

I also like to keep the conversation friendly and non combative :-)

One small point:
The BJT can be modeled as EITHER a voltage controlled current source OR a current controlled current source.  Anyone that says to 'forget' about calling it a current controlled current source is hell bent on world domination or else is just trying to make a point and get others to want to see their point of view (ie the WOW factor which gains more attention).  Sometimes (more often today than in the past) writers use their artistic license to use sarcasm to illustrate a point...i do it once in a while myself.

Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 18, 2017, 02:21:39 pm
I have to admire your patience.  I may have lost mine just a little bit here while you have kept yours intact so i must admire that.

Thank you. Yes - I am trying to remain patient - but there are limits.

Please try to keep the reply down as short as possible as this makes it easier to understand and reply.

Yes - I will do that.

You're saying on the one hand that you believe in BJT voltage controlled, yet you are keeping Vbe constant.

Simply wrong.
Show me one sentence where I propose/suggest to „keep Vbe constant“?
(I hope you do not mean the classical method where for calculation purposes all of us start with a „best guess“ for Vbe).

For example......do you believe that the form Ic=Beta*Ib is useful?

It is not only „useful“ - it is a valid relation (resp. definition) and all of us - including me - use it (as I did in my list of design steps). However, we must not misinterpret this relation.

I think your answer is 'yes' from what you said, and you also said that Vbe doesnt matter.  If Vbe does not matter than why try to push voltage control?

Again simply wrong. For the second time you are using a misquotation (entirely fictitious).
Why? I never have written something like „Vbe doesn`t matter“..

The BJT can be modeled as EITHER a voltage controlled current source OR a current controlled current source.

At least 4 small-signal two-port BJT models are in common use: h-parameters, g-parameters, y-parameters, z-parameters. All four models have a signal voltage as input quantity. However, I am aware that this fact alone cannot be regarded as a final proof  for real voltage control because they are only „models“. Hence, I did not use these models as justification.

I also like to keep the conversation friendly and non combative :-)

Fine - but why do you quote incorrectly ?

Title: Re: Current controlled device vs voltage controlled?
Post by: MrAl on February 18, 2017, 03:08:40 pm
Hi,

"How can VBE determine the current IC when it is (a) not known exactly and (b) assumed to be a constant value ??"

If VBE in your statement is "assumed to be a constant value" then you are stating that it does not change.
How is that a misquote?

Not only that, but i cant see how you can say that the device is "voltage controlled" and ask the question above that implies that you can NOT determine the collector current from VBE.

I hope you are not hinting that just because the base voltage changes that means it is voltage controlled.
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 18, 2017, 04:53:42 pm
Hi,
"How can VBE determine the current IC when it is (a) not known exactly and (b) assumed to be a constant value ??"

Here I repeat what I really have written in my post#59

".....very often I`ve heard the question (not only from beginners):
How can VBE determine the current IC when it is (a) not known exactly and (b) assumed to be a constant value ??"

As everybody can read - I have quoted a question I often have heard from other persons.
And what are you doing? You give the impression as if that were my statement.

OK - no chance for a fair exchange of arguments. I stop the discussion now because you are falsifying my words.
Bye
LvW

Title: Re: Current controlled device vs voltage controlled?
Post by: MrAl on February 19, 2017, 09:12:57 am
Hi,
"How can VBE determine the current IC when it is (a) not known exactly and (b) assumed to be a constant value ??"

Here I repeat what I really have written in my post#59

".....very often I`ve heard the question (not only from beginners):
How can VBE determine the current IC when it is (a) not known exactly and (b) assumed to be a constant value ??"

As everybody can read - I have quoted a question I often have heard from other persons.
And what are you doing? You give the impression as if that were my statement.

OK - no chance for a fair exchange of arguments. I stop the discussion now because you are falsifying my words.
Bye
LvW

Hi,

Well, i have to chuckle a little there :-)

But i do apologize if i misquoted you.  It is harder to talk about something so specific in a message board forum because we can not correct each other right away, it takes sometimes 1/2 day or even a full day before one or more of us can get back to reply.  I've also had a host of other things going on in my court the last week or so that demand a lot of attention so being preoccupied does not help at all.

But be sure i am not trying to 'falsify' your comments, if anything i am trying to verify and understand what you were saying.  However, we all have our way of looking at things and i dont think your view will change much and i dont think my view will change much, and i think the only reason we got into this discussion in the first place was because we dont understand each others points of view entirely.  For example, it's not the first time i have heard the "voltage controlled" idea come up in BJT analysis, and i reject it outright for a number of reasons, and some of those reasons have to do with the history of electronics and how hard it is to change the trend that follows in those footsteps. [Note: of course i meant not voltage controlled only not that we can never use the voltage control idea.]

Because i have studied this problem in detail before, i should have just stated from the start that probably the only way i would accept a 'final' definition of what causes one or the other (voltage or current, or as we say charge) is a proof that there is a *delay* in one when the other somehow appears by itself without the other.  In the context of this discussion, that would mean i would have to see a proof that voltage can appear all by itself, and then sometime later, even if it is such an extremely tiny time period later, that then and only then does the charge start to move (or current starts to flow).  The delay i am talking about here could be on the order of 1e-30 seconds, and even though that is probably not measurable i would still accept that if a good proof was shown.
But to add to that, i would then have to take that argument and *apply* it to a circuit and try to determine if it has any practical significance.  If we can *never* measure that then there is no practical significance, at least probably not in our lifetimes.  At some point technology may be able to measure such an extremely short time, at which case it *could* become significant, but for our time it may not be so.

So to convince me there is only one way that i know of, and that is to show that there is at least sometimes a delay between the time that a voltage appears to the time that charge begins to flow.  Once we have a numerical figure to work with we can then go on to determine the practical significance, if any.  Without that extra step though i could still probably accept that there is at least some time no matter how small, but i would have to at least see that there is some time delay.
Since we are talking about an electrical behavior, we also limit the discussion to electrical behavior alone without the possibility of mechanical or other types of intervention.  A photo transistor will obviously change the properties because of the photo electric effect, but we block all light from reaching the transistor junctions so we can concentrate on the purely electrical (voltage and current) properties alone.

So show me that and i will probably be at least partially convinced.  Of course if that were to happen, we would then have an awful lot of data sheets to change :-)
Since this probably isnt going to happen, i will respect your decision to not discuss this any further.  I realize that it would be very hard to prove this without a good lab unless we could find someone else who did work in this area already, which could be a possibility, and then we would just have to find that work.

BTW an analogy may come in the form of a purely mechanical system.  We have a hard, small ball that is on a frictionless surface and we apply a force to one side of the ball.  What comes first, the force on the ball or the movement of the ball?
Amazingly, with no friction (inertia only) from the moment the force is applied the ball BEGINS to move, and there is no delay in this process.  In fact, i think dynamic friction would not change that either however static friction would because it would hold the ball in position until some significant non zero level of force was achieved.  So again we see no delay between one thing and the other, between the applied force and the movement of the object.

Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 19, 2017, 02:57:30 pm
MrAl, I do not want to be impolite - however, as stated in my last post: In my opinion, it makes no sense to futher continue THIS discussion.
All I can contribute for explaining/justifying my position is contained in my previous posts - in particular posts#7, 27 and 59.
LvW
Title: Re: Current controlled device vs voltage controlled?
Post by: MrAl on February 19, 2017, 04:02:04 pm
MrAl, I do not want to be impolite - however, as stated in my last post: In my opinion, it makes no sense to futher continue THIS discussion.
All I can contribute for explaining/justifying my position is contained in my previous posts - in particular posts#7, 27 and 59.
LvW

Hi again,

Ok sure, no problem.  If i find anything (except my own work) i'll post it here if i think it will help.  Websites with information help sometimes and sometimes not, but if i find anything good i'll post it here at a later time.

Take care for now, and good luck with your future projects/studies.
Title: Re: Current controlled device vs voltage controlled?
Post by: LvW on February 19, 2017, 04:28:08 pm
Take care for now, and good luck with your future projects/studies.
Thank you - same to you.
Regards
LvW