Timeline for Why is a BJT considered "current-controlled"?
Current License: CC BY-SA 3.0
47 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 27, 2016 at 20:22 | comment | added | Incnis Mrsi | Skipped due to unreadable math. | |
| Aug 21, 2016 at 19:06 | comment | added | jbord39 | @LvW: I formally accept your concession, until you respond to the argument i have made over 10 times now. BJT vs. MOS. Flash memory, try it with a BJT you goober. I like how you continually have errors in your @<username>: for responses, so that I do not notice to respond to your ridiculous arguments. | |
| Aug 21, 2016 at 17:02 | comment | added | LvW | jbord39, I think, the niveau of your replies has reached a point where we should stop, OK? | |
| Aug 21, 2016 at 17:00 | comment | added | LvW | @Alfred Centauri, may I ask you to read again the last paragraph of my detailed answer? I think and hope, you can agree to the contents. Or do you have objections? | |
| Aug 21, 2016 at 16:59 | comment | added | jbord39 | @LvW: Very predictable, dogmatic response. Not even reading my answer. You are not arguing with me, becuase nothing you say is related to my points. My point is the BJT is current controlled in comparison the MOSFET which is very clearly voltage controlled. The reason is that a base current is required. This is the distinction. You argue about other nonsense. Go on, keep stroking your 'ego' with big words and 2nd order BJT effects, unrelated to the points made by me. | |
| Aug 21, 2016 at 16:39 | comment | added | LvW | jbord39, sorry to say - but I have the impression you simply do not (or you will not) understand. The base current exist !! But its existance does not mean that it has a controlling function. Hard to understand?. Somewhere you have mentioned "base width modulation". That is a term - nothing else. But WHY is it modulated and WHY does the current goes up? It is always the increasing electrical field (within the decreasing base width) which is responsible for the current increase. Perhaps you now remember the relation between voltage and field? | |
| Aug 21, 2016 at 16:27 | comment | added | jbord39 | @LvW: and to respond to your quote: "Why do textbooks state that BJTs are current controlled". Well, why do they? Becuase a base current is required, in contrast to a MOSFET. You STILL cannot acknowledge this point. BJT's clearly need a base current to conduct. MOSFET's need no such thing. The textbooks refer to BJT's as current controlled IN CONTRAST to the MOSFET. Still, you will not respond to this fact, instead prattling off 2nd order effects unrelated to the base phenomenon (which is that the base must draw current for conduction). | |
| Aug 21, 2016 at 16:24 | comment | added | jbord39 | @LvW: What sort of technical argument is necessary? You have quoted 3-4 sources, including Shockleys original patent at least 5 times. Yet, you cannot come to terms with the fact that there is obviously a requirement for base current, for any collector current to flow. This is a form of control. You are answering a completely different question than the OP is asking, in a very dogmatic, unreasonable way. You admit "the base current is required", but cannot somehow reconcile this with your religious beliefs regarding cause-effect in p-n diodes (which is what you are really arguing about). | |
| Aug 21, 2016 at 16:16 | comment | added | LvW | @jbord39, it is not a fair discussion with you. May I quote the OP? Here is his question: "Why do textbooks state that BJTs are current controlled". As you can see the OP was asking for the working principle of a BJT and NOT for the shown circuit. How can I give a reason when the question contains a wrong assertion? It seems you do not know that SERIOUS and RELIABLE textbooks explain why the BJT is voltage-controlled (Art of Electronics). Your last sentence deserves no further comment. Perhaps a lack of technical arguments? | |
| Aug 21, 2016 at 16:14 | comment | added | Alfred Centauri | @LvW, to be sure, it's not that I'm uninterested in BJT physics, it's that I'm uninterested in the causation debate which I do believe to be pointless. | |
| Aug 21, 2016 at 15:50 | comment | added | jbord39 | What is this argument even for. As I have said time and again, the OP is asking why it is considered. No answer here is going to change the impression that a lot of engineers who don't work with BJT's on a regular basis have (Which is that it IS current controlled). The OP is asking why. Not asking how a BJT works, not asking if it should be considered current controlled. He is asking WHY is it considered current controlled. LvW has repeatedly ignored this distinction on his crusade of ridiculousness (which I can imagine is due to a wrong test question 20+ yrs ago hes still bitter about) | |
| Aug 21, 2016 at 13:49 | comment | added | LvW | @Alfred Centauri, yes - I agree to your last sentence. And that was the reason (you will remember certainly) I wrote earlier: "It depends, if we speak about a circuit with surrounding resistors or about the "naked" transistor". And in your response you was "disappointed" about my "weak" response. Did you not recognize that I was of the same opinion as you ("...what means control") ? Of course, using such a term it is necessary to clarify its meaning. If you are not interested in the BJTs internal physics - no problem. Perhaps a misunderstanding between the two of us? | |
| Aug 21, 2016 at 13:28 | comment | added | Alfred Centauri | @LvW, I'm not interested in engaging in the (what I believe to be pointless) causation debate that you're having with others here. As I wrote in my first comment, there is a sense in which the statement "collector current is controlled by the base current" is true and that if you insist, as you appear to be, that the statement isn't true in any sense, my response is to simply point to the linked simulation. As I wrote earlier, "much of the debate here hinges on what one means by control." | |
| Aug 21, 2016 at 13:01 | comment | added | LvW | @Alfred Centauri, of course, I can do both (appling a current or a voltage). However, as far as I understood the whole discussion, the question is if the shrinking of the depletion area which undoubtly is responsible for the rising of the emitter- resp. collector current is caused (a) by the base current (for my opinion, physically impossible) or (b) by the voltage across the B-E pn junction. That is the reason I cannot accept your simulation as a "proof". Are you aware that the voltage gain of a gain stage does NOT depend on beta (same quiescent point, of course) ? What is your explanation? | |
| Aug 21, 2016 at 12:42 | comment | added | Alfred Centauri | @LvW, it proves, by inspection, that one can control the collector current by controlling the base current. Given the simplicity of the circuit and the definitions of current source and voltage source, I do not know how one can deny this. It's also true one can connect a voltage source across the base-emitter and show that one can control the collector current by controlling the voltage across the base-emitter. The statements are not contradictory. | |
| Aug 21, 2016 at 8:37 | comment | added | LvW | @Alfred Centauri, you have provided a link showing how in a Spice BJT model the output current looks like when there is an ideal current source at the input. Does this proove anything? I am sure you can imagine how the current looks like for a voltage source (0...0.7V) at the input. That is why I call such an "argument" as weak. I am still waiting for a serious justification how a 10µA input current at the base should be able to shrink the depletion layer so much to allow an emitter current of 1mA. Shall I give you more application oriented proofs/justifications? | |
| Aug 21, 2016 at 0:05 | comment | added | Alfred Centauri | @LvW wrote "you qualify my comment as "weak". What does this mean?" - it means that, on my view, you failed to even remotely engage the context of my comment. If you're unable or unwilling to admit that there is a context in which the statement "the base current controls the collector current" is true by inspection, then I must conclude that you've crossed the line dividing the rational from the dogmatic. Also, surely you're aware that if you wish to respond to a comment from a user and desire that they see it, you should use @<username> somewhere in your comment? | |
| Aug 20, 2016 at 16:15 | comment | added | jbord39 | @LvW: blah blah, i know all that, and you are still unable to grasp: the op asked 'why is it considered', not 'should it be considered'. and, we have come full circle (a spiraling loop of degeneration) | |
| Aug 20, 2016 at 16:00 | comment | added | LvW | Yes - that is a good question ("where did it come from?). I think it is because of the simple relation Ic=betaxIb and the (unfortunate) term "current gain". In this context, remember how beta came into the formula: We have started with Ic=(alpha x Ie) and we know (KCL) that Ie=Ic+Ib. Please realize: The factor alpha determines Ic (as the major part of Ie). From this and after some manipulations: Ib=Ic x (1-alpha)/alpha. Hence, Ib is just a small part of Ie. Everything starts with Ie (which is determined by the classical exponential relation between I and V) | |
| Aug 20, 2016 at 15:44 | comment | added | jbord39 | @LvW: Then, where did this whole idea of current-control come from, if no one is under the impression that is how it works? Didn't you quote design fails in analog applications due to this type of thinking? So clearly, the mindset exists. He asks 'why is it considered', not 'should it be considered'. You are like a brick (admittedly with some hard-wired knowledge of analog design) to speak with. | |
| Aug 20, 2016 at 15:38 | comment | added | LvW | "..everyone else does"? Are you sure? Even in this thread you can find votes for pure voltage-control. Do you need serious references - starting with William Shockley (patent document)?? | |
| Aug 20, 2016 at 15:33 | comment | added | jbord39 | @LvW: A "defect" or not, it is an integral part of the working of the BJT. I am not arguing that voltage-control is the proper mindset for analog design (who does real analog design with BJT's these days anyway?? cirrus logic, TI, all use MOS for current designs), but, can acknowledge that the base current is necessary for the forward biasing of the b-e junction. Regardless, as I have said many times, THAT IS NOT THE QUESTION. The question is why is it CONSIDERED current controlled. You don't consider it so, but everyone else does. The reason is the base current. | |
| Aug 20, 2016 at 15:30 | comment | added | jbord39 | @LvW: then you are beyond reasoning with. the model you have married your soul with is just that: a model. progressively more complex models arise as we realize further, deeper interactions. the shockley diode equation is based on other empirical exponential formula, namely the Arrhenius equation. This does not account for the micro level of quantum mechanics, but gives very predictable results (statistics). Alas, it is just a model. Physicists cannot even agree on whether energy is stored in a field; you claiming to have complete understanding of the p-n junction is quite laughable. | |
| Aug 20, 2016 at 15:29 | comment | added | LvW | jboard39 - are you aware that there is not a single proof or justification for current-control? Just assertions - IB exist and, therefore, it has a controlling function. That`s the logic of some guys. Somebody even wrote: "It controls because it matters". If you are really interested I can list you various effects and technical observations which proof that VBE alone is the controlling quantity. IB is just a by-product (Barrie Gilbert: "A defect"). That is no surprise - all parts (active or passive) have undesired by-products. | |
| Aug 20, 2016 at 15:20 | comment | added | LvW | jboard39, sorry but I cannot. At first - it is, of course, not a chicken-egg problem. No current without driving voltage. Consider a simple pn junction. Voltage is always a PRECONDITION for a current. Current is the RESULT of an applied voltage. Secondly, I cannot "remove" the current from a conductive body when a voltage is applied. Can you? This is no serious argument. Of course, there is a base current - have I ever denied this fact? But it is just the result of "splitting" the emitter current into a small part (IB) and a larger part (IC). Hence IE=IB+IC. | |
| Aug 20, 2016 at 14:46 | comment | added | jbord39 | @LvW: whats the point dude. I really don't understand. It's like you got this answer wrong a test 25 years ago and never let it go. Clearly you need base current for the BJT to work. It is "controlled" by base current. Is it controlled by voltage too? Hell yeah. Which one controls it more? I don't care, which comes first the chicken or egg? How can you argue that the base current will not control the collector current? Sure, a voltage is involved and may even be the primary driver, but remove either and you get a cutoff BJT. Can you at least agree to this last statement? | |
| Aug 20, 2016 at 7:46 | comment | added | LvW | AlfredCentauri - to be honest, I only have tried to be polite. You qualify my comment as "weak". What does this mean? Do not hesitate to tell me where I am wrong - can you? Do you really think that your simulation with an ideal current source at the base can proof anything (THIS is really "weak")? In the Spice model, such a current creates a voltage Vbe which does the job. Only a voltage can cause a reduction of the depletion area - a precondition for a current. I cannot believe that these pn junction basics are new for you. | |
| Aug 20, 2016 at 1:10 | comment | added | Alfred Centauri | @LvW, I am disappointed in, though not entirely surprised at, your very weak response. | |
| Aug 19, 2016 at 7:03 | comment | added | LvW | Alfred Centauri - dont mix the behaviour of a circuit (BJT plus other parts) with the working principle of the BJT alone. Of course, I can use an external voltage source (and a large series resistor) - and that`s what we call "current source" - for producing the voltage VBE necessary to open and control the BJT (in fact: to open ans steer the depletion layer). See my example with an opamp wired as a current-to-voltage converter. | |
| Aug 18, 2016 at 20:57 | comment | added | Sredni Vashtar | @horta actually there is a third party at play, here. The agnostic party. Even Mosfets have gate currents... | |
| Aug 18, 2016 at 20:31 | comment | added | Alfred Centauri | I suspect that much of the debate here hinges on what one means by control. Since a simple SPICE simulation will confirm that one can control the collector current by controlling the base current, the statement "collector current is controlled by base current" is indisputably true in that sense. If, like LvW and wbeaty appear to, one chooses to insist that such a statement is false in any sense, I will simply point to this: i.sstatic.net/LqFx1.png | |
| Aug 18, 2016 at 19:15 | comment | added | horta | Since voltage and current in devices can't exist without each other, this is a religious war. This answer just adds more gas to this religious war. Current control or voltage control is simply hand-waving to reach an approximation or general engineering understanding quicker. | |
| Aug 18, 2016 at 18:43 | comment | added | jbord39 | @wbeaty: what's ridiculous is that I agree with both LvW and you, on how BJT's work. Yet, I can still understand that the BJT requires current to function. Plus, I feel like the V-I dichotomy is just a duality in this case, as shown by taking the natural log of the shockley diode equation. But, I guess two slightly opposing thoughts are too much for your head to handle (with all that theory packed in there!!) | |
| Aug 18, 2016 at 18:40 | comment | added | jbord39 | @wbeaty: But, can you run a BJT with zero base current? Is a base current absolutely necessary for the BJT to function? yes, it is. With no base current, the BJT is cut off. This whole small signal stuff is just window dressing which distracts from the actual question. I seriously don't understand how you cannot grasp this simple distinction in comparison to a MOSFET (which needs no constant gate current to remain conducting: which gives rise to things like flash memory, with charge stored on the insulating gate keeping the MOS permanently conducting). | |
| Aug 18, 2016 at 18:34 | comment | added | wbeaty | OF COURSE IT'S NOT A RELIGION, instead it's physics/engineers versus the incorrect beliefs taught in grade-school. We're supposed to give up those simple models when we get to higher levels (undergrad EE.) Diff amps cannot be explained by hfe-based models. Neither can current mirrors. Neither can cascode amps. IMPORTANT: if you believe that ib controlls Ic, then for you modern audio amps will be forever behind a barrier of confusion and ignorance, since DC-coupled audio circuits use voltage-based BJT designs where hfe is irrelevant. Similar situation: look at the interior of TL071 etc. | |
| Aug 18, 2016 at 17:36 | comment | added | LvW | Yes - perhaps you are right. But I try to understand the following phenomenon: There is not a single technical/physical justification for current-control. Nevertheless - some people are ignoring all proofs in favour for voltage control. They simply claim that there is a base current (which is, of course, true) and that, therefore, this current would have a controlling function - opposite to many observable effects and circuit properties which can be explained with voltage control only. Is transistor technique a matter of faith? | |
| Aug 18, 2016 at 17:07 | comment | added | George Herold | I'm upvoting, but LvW you seem to have a bee in your bonnet when it comes to this question. Sometimes it's useful to think of a bipolar as a current controlled device. (Setting the bias resistors for linear operation, choosing R value for capacitance multiplier. I don't use single transistors much.) Other times it's not useful. | |
| Aug 18, 2016 at 16:01 | comment | added | LvW | Sorry - you are wrong. Do you know which model is used in all circuit simulation programs? Did you ask yourself WHY we are allowed to assume a certain voltage VBE (example: 0.65V) for designing an amplifier stage (answer: voltage feedback!).? Your equation is correct - however, it is used for finding the resistor Rb=(Vin-Vbe)/Ib for producing the required voltage Vbe using the necessary voltage drop IbRb. Ib does exist - no question about it. But this does not mean that it has a controlling function! Remember: The voltage gain of a transistor stage does NOT depend on B or beta! | |
| Aug 18, 2016 at 15:27 | comment | added | Sredni Vashtar | That simplified model can be used to design the simple BJT circuit shown by the OP. As a matter of fact it's that simplified model that is used when you assume Vbe = 0.65 and compute Ib = (Vin-Vbe)/Rb. Also, I believe that Leon O. Chua, Charles A. Desoer and Ernest S. Kuh have heard of Maxwell and Boltzmann, too. The point is that the map is not the territory. There are so many models: Ebers Moll is just one of them; Gummel Poon in another, more complex; the one I mentioned is yet another, much simpler. But it shows why the BJT can be considered current-controlled. | |
| Aug 18, 2016 at 14:45 | comment | added | LvW | Do you consider your comment really as an attempt to disproove something? I cannot believe! You simplify the I-V characteristic by SETTING V=const - and as the next step, you state that in your model the current is independent on V. Surprising logic - a circulus vitiosus? I suppose you have heard about Shockley`s exponenetial I-V characteristic for a pn junction? | |
| Aug 18, 2016 at 14:17 | comment | added | Sredni Vashtar | @LvW , my comment above counts as one attempt to disprove your statement. You can find the simplified model on page 141 of Linear and Nonlinear Circuits (Chua, Desoer, Kuh) | |
| Aug 18, 2016 at 14:06 | history | edited | LvW | CC BY-SA 3.0 |
added 55 characters in body
|
| Aug 18, 2016 at 14:05 | comment | added | LvW | Yes - that´s what I have expected: The downvoting ! This confirms my conviction: For some people, this question is just a matter of faith - nothing else. No serious technical questions, no counter arguments, not a single attempt to disproove one of my statemets. Not a good sign ! | |
| Aug 18, 2016 at 13:57 | comment | added | LvW | jboard39, did you read somewhere in my answer that Ib=0? Read again carefully! | |
| Aug 18, 2016 at 13:44 | comment | added | Sredni Vashtar | Ok, let's try this way. Use a simplified diode model, with a piecewise linear charateristic: I=0 for V<Vth and vertical V-I characteristic for V>=Vth. Use this model for the B-E diode: that means Vbe has a constant value. How do you voltage control that? | |
| Aug 18, 2016 at 11:59 | comment | added | jbord39 | Try driving a BJT base without current (bias or otherwise) and tell me how it works out for you buddy. Alternatively a MOSFET draws no gate bias current. Hence the distinction. | |
| Aug 18, 2016 at 10:58 | history | answered | LvW | CC BY-SA 3.0 |
