Current Feedback Amplifiers, not only a semantic problem?

Jump to Latest
Smiley 09

Smiley 09

Member
Joined 2009
Bonsai

Bonsai

www.hifisonix.com
Joined 2003
hifisonix.com
Paid Member
As I remarked many posts back, if you want to prove CFA = VFA then just raise Rf and Rg by a factor of 10 and see if the CFA still works. It doesn’t. Crude, but it proves the point that you need a strong measure of current flowing in the feedback network for correct operation. I still prefer to use my original definition of a CFA: peak current into the TIS/TAS node is determined by the value Rf.

(Scott: Separately, I reran my H-bridge simulation as I could not find my original hard copy plot that I posted up on the original CFA vs VFA thread. Indeed it is VFA. I fell into the same trap Inpointed out to Herve in the independent gain/bandwidth behsviour at low(er) closed loop gain levels was a result of sub-optimal compensation. The amp in the model would have been useless in practice).
 
jan.didden

jan.didden

AX tech editor
Joined 2002
linearaudio.net
Paid Member
H

Hans Polak

Member
Joined 2005
We have had interesting discussions with some sidesteps, but now have come to a point without any further signs of progression.
From here on this discussion could go on for ever, like discussions concerning our climate control and "never the twain shall meet".

If we are brave enough we could kill the semantics issue with the sound of a "fake" gunshot :D and agree that the name CFA was given to a specific topology, as already mentioned more than once.

A specific and very simple test confirming this CFA topology, separating it from a VFA, has been proposed and there were no objections against this test as far as I know.


Hans
 
forr

forr

Member
Joined 2004
syn08 said:
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.
Click to expand...
The often used definition of CFA by its well understood properties does not avoid misconceptions in its intimate way of working which does not differ from the VFA one: the control of the feedback loop is anyway under the control of the transconductance of an active device. The CFA naming inevitably induces errors in the interpretation. I do not think it is too late to change the naming of CFA. Can't the creativity in expressions of the marketing services give a little help ?
 
Bonsai

Bonsai

www.hifisonix.com
Joined 2003
hifisonix.com
Paid Member
Hans Polak said:
We have had interesting discussions with some sidesteps, but now have come to a point without any further signs of progression.
From here on this discussion could go on for ever, like discussions concerning our climate control and "never the twain shall meet".

If we are brave enough we could kill the semantics issue with the sound of a "fake" gunshot :D and agree that the name CFA was given to a specific topology, as already mentioned more than once.

A specific and very simple test confirming this CFA topology, separating it from a VFA, has been proposed and there were no objections against this test as far as I know.


Hans
Click to expand...

I agree with most of this. However, I think you should just be able to look at a circuit (lets include the component values just for completeness) and say its one or the other. We should not have to resort to a sim IMV to determine topology. But, that's only my view and I am happy to be corrected or adopt the general consensus on this one.
 
C

CPaul

Member
Joined 2011
IanHegglun said:
??? 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)
Click to expand...

Not exactly sure what you're asserting here. But I spoke with Sergio. He's never posted to Wikipedia.
 
Max Headroom

Max Headroom

Member
Joined 2013
From what I see, VFA employs differential voltage stage to derive ops drive signal voltage, CFA (effectively) employs feedback energy directly to the input energy point to control the derived ops drive signal voltage.
These error correction/cancellation modes are fundamentally different and have quite different noise behaviour implications.
VSSA ( VFA vs CFA )
As expected, the CFA is here 5 time faster than the VFA. And this is not due to the miller cap, but the input stage topology: the way the feedback signals are substracted from the input ones.
This is due to the two input stage topologies transconductances: The Long-tail Pair is "compressive" while the CFA is "expansive" (© Richard Marsh).
In other words, think about the "current on demand" behavior of the CFAs, too well described in numerous papers to be pontificated here ;-)
Click to expand...
What is meant objectively and/or subjectively by the terms 'compressive' and 'expansive' ?.


Dan.
 
Bonsai

Bonsai

www.hifisonix.com
Joined 2003
hifisonix.com
Paid Member
He used those terms to differentiate the behaviour of a VFA vss aa CFA as follows:-
In a current source fed LTP (the most common configuration), the current is 'steered' into or away from the TIS input. Its peak value can never be more than the LTP current source, or 2x the standing current for each LTP half. In a CFA, the current on demand (COD) behaviour means the peak current into the TIS/TAS can be much higher than 2x the standing current as is the case in a VFA. You can sim this and in a properly compensated CFA, get 5-8x the standing current.

Note the above comments refer to 'classic' VFA and CFA. There are optional compensation techniques in VFA and circuit tweaks that can also emulate CFA COD behaviour.
 
Last edited:
Bonsai

Bonsai

www.hifisonix.com
Joined 2003
hifisonix.com
Paid Member
forr said:
The often used definition of CFA by its well understood properties does not avoid misconceptions in its intimate way of working which does not differ from the VFA one: the control of the feedback loop is anyway under the control of the transconductance of an active device. The CFA naming inevitably induces errors in the interpretation. I do not think it is too late to change the naming of CFA. Can't the creativity in expressions of the marketing services give a little help ?
Click to expand...

COA= current output amplifier. The output controlled quantity is a current. You can make this with a 'CFA' or a VFA building block

CFA = an amplifier that represents the output controlled quantity to its inverting input using a representative current. The output controlled quantity of such an amplifier can be a voltage or a current.

So, it is the name of the 'current output amplifier' that needs to be made clear. It never was a 'current feedback amplifier' - that is a ridiculous name for it.

The CFA name fits perfectly with how the circuit works.
 
You must log in or register to reply here.
Top Bottom