Do you still believe the current flows from the output to the emitter input in the circuit with no Rg? Your next comment shows you still believe this. The direction is obvious. So I see little point in continuing this dialogue.
In a previous post you wrote (post 1866)
The single ended circuit with no Rg is the one that defines the underlying feedback principle of the CFA. My analysis shows no current feedback only voltage feedback from the output to the input stage, a bootstrapping effect.
As for the feed forward question this was mentioned in post 2074 Current Feedback Amplifiers, not only a semantic problem?
for the same circuit but with Rf and Rg where Middlebrook looks at ways to account for this feed forward current with his DIT test. Link http://www.intusoft.com/lit/GFTManual.pdf
Did you notice he calls this circuit "a typical 2-stage voltage-to-voltage feedback amplifier"? And presumably this was written before the CFA name was introduced (either that or Middlebrook rejected the CFA name - does anyone here know?).
Yes, the term "Current Feedback Amplifier" suddenly appeared in a patent and the examiner obviously was not able to determine it's validity (or not) on technical grounds (look at the analysis 40 years later and still in dispute!). So CFA was accepted in a legal document. Any company wishing to contest the validity of the term CFA would be faced with court costs in the millions to hundreds of millions of dollars (eg Kodak). Looks like the battle was not worth the cost.
Now the original patent has expired there are likely newer ones using the phrase "current feedback" or referencing the original so it likely remains protected under patent law. And those working in the field are not likely to go against the grain and speak out about it as being a misnomer, fearing loss of reputation or their job, and you only get insiders who try to support the status quo.
Cheers,
??? check the references. Both are Sergio Franco. The style circuit in Wikipedia match. The use of the phrase "...CFA is a type of electronic amplifier whose inverting input is sensitive to current..." has been used in Sergio's writings.
I don't agree with the above explanation and I think it should be removed.
Has any diyAudio members contributed to this Wikipedia page? And is willing to try?
Cheers,
Ian Hegglun
https://en.wikipedia.org/wiki/Current_(electricity)
Patent law has nothing to do with these matters, patents are full of nonsense terms and flawed descriptions. All that is protected is what is specified in the claims. I'll remind folks that Nelson's patents did not claim the underlying principles (they were prior art) just his specific embodiment. The IC industry never had any concern over integrating the concept (AD844, etc.) anyone who thinks there were any royalties or concessions in this matter is sorely mistaken.
Hi Scott,
I have no idea whether you are still working for the industry. I respectfully ask if you free to say what you like now? Some reassurance.
Cheers,
The first were discrete and IIRC the biasing and some other details were what were actually patented. As I said backgrounds and descriptions are not what is patented. They are often full of the most horrifying audiophile hubris (in the case of some audio amplifier patents) there is no standard or check and balance in place for this part of a patent.
No I am 100% retired (I remain respectful of all those I have worked with) but I don't get what you mean. I think this stuff about where the current goes at gain of 1 and DC is not relevant to anything, if that's what you mean. Virtually everyone I have worked with in the industry for 40yr. would laugh at the last couple of weeks of this thread.
Hi Scott,
I don't know if it against this forums rules to question members' motive(s) -- in this post it related to what you say here in relation to CFA being a misnomer (the subject of this thread).
We all use discernment when interpreting what others say and why they say that and refrain from saying other things. If you are free to say what you believe then that would help me for one to accept what you say on face value rather than possibly doubt it due to ulterior motives.
BTW I heard an ex-bank executive who is now speaking freely about problems he saw and he said he waited 10 years after leaving the industry before speaking out because he considered that to be the 'ethical' waiting time.
Cheers,
Last edited:
There are much more independant people than I thought who are critical of both the semantics and the interpretation of the CFA concept. I found instructive posts in this thread
CFA Topology Audio Amplifiers
Here are a few excerpts I think to be meaningful.
Last edited:
Are you serious? Do you think there is some kind of industry "cover-up" where we all protect each other. This is the stupidest thing I've seen in years. Do you actually think anyone cares about this?
Explanation in the shortest form how works a Sziklai pair where it is clear that the emitter of the input transistor is an ouput :
I guess our first problem is that our herd of cats can't even agree on how the darn thing works. So how are we supposed to choose a name for it?
But seriously, I can't find a definition of "Current Mode Amplifier." I believe I read a paper from Barrie Gilbert talking about designing with current rather than voltage transport modes, but this doesn't address feedback. And of course we have that unfortunate historical precedent of "current feedback" covering the case where a VFA gain block accepts feedback from the voltage across a current sense resistor in series with a load. I think that this is the basis for some of the resistance to the name and description of a CFA.
But I don't know where this leaves us all.
Middlebrook's DIT has the answer. Not as to the topology of a circuit, but as to the amounts of current and voltage loop gains. The lesser of the two has the greater amount of feedback.
Scott, I believe you are evaluating what’s written here with the wrong metric. It is obvious nobody in the industry has the time and the mood to debate how many angels can dance on the CFA pin head. As long as the CFA properties are well understood, and everybody and their mothers understand what’s about when the CFA concept is mentioned, everything is fine and will remain so for the foreseeable future, as you say, who gives a **** about the pesky details.
Otherwise, this forum is not really the industry, so questioning the naming convention defined in the early 80’s for that ******* circuit topology otherwise known from before WWII is, why not, legitimate. As I said before, the only surprise I had in this endless discussion was the prof. Sergio Franco hard stance and (IMO) wrong definition and interpretation. Somebody coming from the academia should know better, but then it would not be the first time when engineers are beating a famous professor with a stick. Watch this, if you have the time:
YouTube
FWIW, I had last year a short email exchange with prof. Sansen from KU Leuwen (he was one of my mentors back in my happy PhD student days) and the CFA topic came up; he absolutely agrees that the “current feedback” name was abused in CFA, but then he rightfully (IMO) thinks in terms of “so what”. It’s just a naming convention universally accepted, as much (my comment) as we recognize and accept in an action movie the sound of a gun shot as “real”, although it is not even close to the real sound of a 45. And nobody complains that the sound track is wrong
.Speaking about CFA audio applications, I’m having though a very hard time understanding the stubbornness of both Bob Cordell and Doug Self to snub CFAs in their otherwise excellent (and IMO kind of matching PNP-NPN pair) books.
If there is anything I find really frustrating in this thread is the lack of understanding of the small vs. large signal analysis and models. But then, I suppose it’s part of the learning curve for some
.
Last edited:
I'm afraid that once again, you are making incorrect inferences.
Your analysis still avoids the application of DIT, which would show the relative amounts of loop gains and feedbacks.
Frankly, I don't care what he called it. I do care what his DIT measurement technique shows about it. Why don't you? Run the darn thing and explain its results.
The patent examiner's analysis is in dispute? I'm referring to how the claimant described it. That you can find disputants for almost anything is not surprising.
You're kidding, right? The term CFA wasn't patented. The design was! And yet multiple companies are manufacturing CFAs and seem completely unconcerned about patent violations of any kind. And even if they were, no one patented explanations of how the circuit works.
Bottom line - nothing is holding any of them back from explaining how these ICs "really" work. And yet no one does.
The phrase "current feedback" could not possibly have been patented. It was in use in the 1940's in the Radiotron Designer's Handbook. And even if it were patented, all the more reason to invent a new term and a new description of how something works to avoid the patent.
Honestly, this sounds to me as if you're talking of conspiracy theories.
Bringing up fake arguments doesn’t help the cause; it was only mentioned in the classic sense of output current sampling.
I’ve posted earlier the link to the document, everybody with an unbiased open mind can recognize the truth.
Otherwise, your very own DIT based definition of “current feedback” is nothing but laughable. Nobody with EE understanding of feedback theory is going to argue about it, you cannot tie an alleged general definition to a particular analysis method. What was “current feedback” before Middlebrook proved his theorem?
Last edited:
Ian H: I agree with Scott. I cannot think of any reason why the actual direction of the feedback current has anything to do whether it is or is not CFB. The direction, imnsho, is an interplay of input signal, loop gain, feedback factor and probably a few more things.
Jan
Last edited:
Last edited: