As every transistor working in usual conditions. But what does control, drive, steer (or any adequate verb) the emitter current ?
Current-mode
By the way, some people may not know it, so I recall that in the discussion here or maybe in AudioXpress (see post #1 for hte links), there was a reference to an article by Barry Gilbert (who supports the CFA concept) :
(PDF) The current-mode muddle
A short quote (at the beginning of chapter 2) to incite to read it :
It is by now clear that what is popularly called a 'current mode' circuit may not fully justify the name.
By the way, some people may not know it, so I recall that in the discussion here or maybe in AudioXpress (see post #1 for hte links), there was a reference to an article by Barry Gilbert (who supports the CFA concept) :
(PDF) The current-mode muddle
A short quote (at the beginning of chapter 2) to incite to read it :
It is by now clear that what is popularly called a 'current mode' circuit may not fully justify the name.
No, because if you change Rg, you change the open loop gain, so the OLG will change.
Why not thinking there is space between pure current feedback and pure voltage feedback ?
Last edited:
You used 'open loop gain' and 'OLG' which I believe are the same. Is this a typo?
BTW I do agree that there's a gray area between pure VFB and pure CFB if your criterium is the input impedance of the inv input. If your criterium is the way the output current of the input device is developed, I don't think there is a gray area.
Jan
Last edited:
I have seen too many times the common ploy of refusing to address an inconvenient truth by attempting to switch attention to a different challenge.
Sorry. I'm not falling for it this time.
Address the claims of 956 first. Then we can talk about something else.
The answer to "Will you?" is, so far at least, you haven't.
I do not believe we should be arguing from authority or teasing others with reading assignments that they would have to undertake to try to first identify, and then to understand, some point that is as yet unknown.
I would be happy, however, to hear a summary of the argument that you wish to present. I can then choose to read the reference or not.
Last edited:
I've never seen that formulation.
We have to be careful with our definitions. Conservation of current demands that in the three terminal transistor, ie = ib + ic.
A nice little summary of various small signal models can be found at BJTs after Biasing: Analyzing BJTs with a Small-Signal Model It uses the terms beta, alpha, RE, gm, rpi and Ross to state a number of relationships between ie, ic, ib and vbe.
A much more detailed discussion can be found at http://alan.ece.gatech.edu/ECE3040/Lectures/Lecture20-BJT Small Signal Model.pdf.
I think it is useful to ask another question. Did the engineering community miss something? Were there off the shelf solutions for all these applications all along? Please substantiate that it's the same thing under a new name.
The same question applies to the Demrow instrumentation amplifier (AD524, INA 103, etc.). Bringing out all the internal nodes and asking customers to create their own circuits is not a good answer, these types of parts have very limited commercial appeal.
You need to be careful there is a small but vocal group (including folks that should know better) that think beta and forcing base current is the fundamental operating principle of BJT's.
Beta and ib are just another inconvenient truth that cannot be denied. I am not willing to designate any fact about transistors as being either fundamental or non-fundamental.
But I think we can both agree that although we must be aware of beta, it is so subject to variation with temperature, emitter DC current and production lot that it would be foolish to base
a design on the expectation that a transistor would have a specific narrow range of beta values.alpha is narrowly constrained to be just a little less than 1 for almost all transistors. And so no transistor can avoid the fact that almost all electrons flowing through the emitter flow through the collector, and all collector electrons flow through the emitter.
I believe that your point is that this fact is inadequate to explain transistor operation, and I agree.
Last edited:
You are right, i miss interpreted the ic=α.ie .The α is to correct ie for the part that doesn't go to ic but to the base. So ib is already acounted for.Originally Posted by Ketje
ic+ib = α.ie ?
I've never seen that formulation.
We have to be careful with our definitions. Conservation of current demands that in the three terminal transistor, ie = ib + ic.http://
Mona
I have replicated and tested your circuit.
You will note that below 1kHz, only about 1nA flows through R9 and R10. Even if all that current flowed through R1 and R3, the voltages at aaa0 and aaa1 would be 1mV – (1nA x 100 ohms) = .9999mV. Of course the voltages are equal!
Now let’s test whether it is more accurate to describe the circuit as a transresistance amplifier or a voltage amplifier.
As you can see, despite the fact that the simulator is having some numeric issues, in this case, the transresistance model is more accurate.
Now please, no more diversions until you comment on post 956.
I was unaware that prokaroytic microbes were experimenting with audio. Who knew?
Strange, Jan - I can. Thanks for letting me know. Here it is embedded in case anyone else is having the same problem.
Attachments
I have neither read nor written anything which could look like such a denial.
I = Is exp (vbe /(kT)) is constantly reminded by those who deny the physical possibility of the principle of current feedack because they have not yet received a satisfactoy answer to how ie is defined in the process.
Since they have not responded to threads that answer it, it is quite possible that the explanation is that they haven't received it for some reason.
Recall the simple load-less CFA, biased by only a 1mA DC current source? Where you agreed that for AC currents, ie1 = ic2?
Input stage AC current is output stage AC current. I cannot image how this can be described as anything other than current feedback. And yet, the physics of the transistors haven't changed; it is still true that I = Is exp (vbe /(kT)).
Current feedback co-exists quite nicely with I = Is exp (vbe /(kT)).
I thought of Bod Widlar when I read your articles in AudioXpress. My post was a slight, not unuseful, digression as it can inform some readers that there are more than one way to make high speed VFAs.
Last edited:
They converge on Rg.
Input stage DC and AC current is under the control of a voltage.=^)
So there is only one more question.
Is it the feedback current which determines Vbe or Vbe which determines the feedback current ?
This is as meaningless a question as asking whether a current through a resistor causes a voltage across it or a voltage across it causes a current through it.
vbe is some one-to-one function of ie and ie is that inverse function of vbe. Therefore, ic is also some function of ie as well as a function of vbe. You cannot divorce the voltage from the current simply because one relationship might be mathematically more tractable and more convenient to express.
The refusal to acknowledge this voltage / current duality continues to surprise me.
Let us now dissect language. In the circuit we are discussing, the output stage current feeds into the input stage. Since the signal exciting the circuit affects the input stage at an earlier instant than the output stage, this "feed" is "back" in time. Hence, current feed back. The relationship of ic to vbe does not contravene this fact.
It is also worth noting that in this and in a number of other circuits, the value of gm has practically no effect on circuit operation - other than to establish a value of vbe.
Last edited: