I did not say you did not respond to my posts but that you or anybody else did not dismiss my approach.Originally Posted by forr
I point out that my analysis considering the effect of the load of the inverting input has not been dismissed.
If you'd like, please indicate the post that was not responded to. It's not that I wish to continue to argue, but that I have tried to make a point of not ignoring what others are saying.
May I precise that sometimes the american language can escape to me and I may fail to understand some sentences. I said about your approach of CFA that your schematics are mainly conceptual and less explicit than mine. You had the contrary feeling. This is where a third party should have been useful, some people able to understand schematics but not aware of the CFA/VFA controversy. They could have asked questions and give their opinion on the whole, the explantions and the concept. Unhappily there were not such artless interventions.
I think the disagreement comes from the determination of the ultimate phenomenon inside the input stage which allows the feedback to have its regulating effect.
A further note : the refutation or the acquiescence of the CFA concept does not prevent the fact that there are great designers on both sides. Most probably, more on the CFA side. But the CFA is not a matter which can be settled by a poll.
Last edited:
Would you care to reacquaint us with your approach?
Surely post 1028 is quite specific rather than conceptual. Can't you comment on that?
I expect that we all share a frustration: how can so many of us who are clearly fairly well acquainted with the functioning of circuits, fail to come to an agreement on such a basic question?
I recognize your objection to the use of a term which once was the name of one kind of a circuit and is now the name of another. (Personally, it does not bother me much, because I think the term was poorly applied in the first place.) Can this be the major basis for our disagreement?
Is the disagreement as simple as one based on definitions? For instance, if in your view, terms such as "current feedback" should be reserved to describe a function that the circuit seeks to accomplish (control the current through a load), then it cannot be used to describe how the circuit functions. But then, neither can we use the term "voltage feedback" to describe how a circuit functions, because it must be reserved to describe a circuit which seeks to control the voltage across a load. In the end, this approach leaves us with a very meager vocabulary for describing circuit operation.
So if a current equal to the difference between two currents, one from a CFA output and one flowing into ground, feeds into the emitter of a CFA input stage, and 99% or more of that current flows into the input stage's collector, the current from which is proportional to the output voltage, why is this not a feedback of current?
In discussions like this it is easy to get bogged down in detailed and elaborated schematics.
The best way to make clear the various positions is exactly with conceptual diagrams. If you can't convince your audience from a conceptual diagram, no way you can do it from a detailed schematic.
If you can't 'win' with the concept, you're most probably seeing it wrong.
Jan
The best way to make clear the various positions is exactly with conceptual diagrams. If you can't convince your audience from a conceptual diagram, no way you can do it from a detailed schematic.
If you can't 'win' with the concept, you're most probably seeing it wrong.
Jan
The thread title is "Current Feedback Amplifiers, not only a semantic problem?"
Now we know that the semantic part of the "problem" is shocking, at least to a minority.
However, the thread title asks if there is at least one other problem with CFA. I wonder what it could be? My AD844's work like expected, and well. I expect eg Bonsai's designs, or Nelsons F5, do also work without "problems", and even very well.
Does someone care to elaborate what that other problem could be, after 1063 posts about that shocking semantic problem?
Now we know that the semantic part of the "problem" is shocking, at least to a minority.
However, the thread title asks if there is at least one other problem with CFA. I wonder what it could be? My AD844's work like expected, and well. I expect eg Bonsai's designs, or Nelsons F5, do also work without "problems", and even very well.
Does someone care to elaborate what that other problem could be, after 1063 posts about that shocking semantic problem?
I'm taken back to kindergarten. This "my amp sounds better than your amp" routine is juvenile and stupid.
That baby was starved of oxygen.
The full quote I can find from Jan is:
Of course, with this logic, we're all seeing it wrong! No one has convinced this thread's audience.
Some will complain that certain arguments are "mainly conceptual" and do not respond to them. But when confronted with a simulation of an "explicit" datasheet's simplified schematic of a commercial CFA, there is (at least as yet) still no response.
Let me offer another adage: If you ignore and refuse to face arguments that contradict your beliefs, there's most likely a problem with your beliefs.
Chris,
I found Sergio Franco's article in EDN most convincing, although I didn't have to be convinced from the beginning.
Your reaction is almost as if it was a direct attack on you, it was not at all.
But I must say, belief and arguments are two worlds that will never meet.
I prefer Jan's definition to keep things conceptual like Sergio Franco did.
Hans
Hi Hans, I quoted your post because I could not find a post where Jan said exactly what you quoted. Perhaps I missed it?
The rest of my remarks were not directed at you; I apologize if it seemed that way.
I don't take anything here personally. But I do reiterate that if someone (not you - review the past few post pages if you want some idea of who) chooses to engage in a disagreement, then it is at the very least impolite to simply and repeatedly ignore someone else's challenges to one's point of view. And the lack of a response itself does strongly suggest why replies are missing.
I have had extensive private discussions with Sergio Franco regarding attacks on his and Walt Jung's work (and on their personal competence too) by Moo Koo. These led in part to my AudioXpress article Current Feedback: Fake News or the Real Deal? | audioXpress . My views of the subject of this thread align very well with Dr. Franco's. If you think it appears otherwise, I'd appreciate you letting me know.
Again, apologies.
I also enjoyed reading Sergio Franco's article in AX. As for MK, off with the fairies for the most part in my view. I read Sergio Franco's articles on CFA's in the 1980's IIRC (EDN) - I remember looking at it and asking myself 'how the Dickens does that work?'. I managed to work through it and grasp the principle and never looked back. I design with both CFA (power amps) and VFA (small signal low noise stuff).
jk n
Begining to teach amplifying electronics by studying a single solid state device is certainly more beneficial for a student than by displaying conceptual schemes. That's why I concentrate on simple circuits with "real" compoments for the fundamental points to investigate what is in under the CFA concept. (post #785). I apologize to CPaul, he effectively showed non-conceptual circuits, post #789 for instance.
I reminds me how as music is often tought in France. Is it better to begin by studying music theory ("solfège" is the french word) or by playing instruments knowing nothing but with an attention to harmony.
Begining to teach amplifying electronics by studying a single solid state device is certainly more beneficial for a student than by displaying conceptual schemes. That's why I concentrate on simple circuits with "real" compoments for the fundamental points to investigate what is in under the CFA concept. (post #785). I apologize to CPaul, he effectively showed non-conceptual circuits, post #789 for instance.
Good question.
There is no doubt that CFA circuits work and may be the best in some applications.
The semantic problem is in two parts.
The first was about the change of meaning around 1975. We can leave it aside by now
The second is subtle, because the meaning is linked to a technical aspect
CF et VF refer to the intimate feedback process, more exactly to what electrical value is the determining factor of the output current of the input stage.
Maybe the center of the discussion could summarized by this question :
is it legitimate to say that an amplifying circuit having a low impedance inverting input is submitted to current feedback ?
I assert that it might or might not be legitimate.
Drive the non-inverting input of a VFA with a voltage source. Connect a low resistance r between the two inputs. As Rg is varied over the range where r < Rg || Rf, there will be a CFA-like constant bandwidth, but no CFA-like slew rate enhancement. Certainly the topology of the VFA is not CFA-like. Since no significant current enters the VFA gain block, I wouldn't call this current feedback.
This was the rebuttal to MK's article some months later (so in 2017 IIRC) in EDN - sorry, it was NOT in AX - my error! The other anti CFA protagonist in the comments section was Kevin Aylward.
But Franco was already writing about CFA's in the 1980's in EDN - long time ago for me but I think the article(s) I am referring to were '88 or '89.
Last edited:
I'm pretty sure Franco never published in aXp. I published a rebuttal to MK in aXp, though. Maybe you're thinking of that?