Current feedback - not suitable for audio ?

Status
Not open for further replies.
Jump to Latest
A

ash_dac

Member
Joined 2005
#1
Hi,

I have been looking at the Accuphase amplifier design, and reading chapter 9 of http://focus.ti.com/lit/an/slod006b/slod006b.pdf

CFA's disadvantages:-

-poor precision
-poor common mode rejection (due to mismatched input impedances)
-poor powersupply rejection
-difficult to apply (stray capacitance..)

CFA's advantages:-

-high slew rate
-wide bandwidth

Has anyone listened to an amplifier that uses current feedback in the design?

Do the advantages of increased slew rate, and bandwidth outweigh the disadvantages ?
 
#3
Do the advantages of increased slew rate, and bandwidth outweigh the disadvantages ?
Click to expand...
You can answer your own question. What slew rate and bandwidth do you require? Then make two designs that meet this, look at the other critical aspects and choose the better one.
The topology doesn't really define anything...it just creates a different set of problems to solve. You might ask which feedback method is "easiest" to achieve necessary performance. But again the answer depends on the skill of the designer. Same with FETs, valves, bipolars and so on.
 
#5
There already were some discussions at DiyAudio about the so called "current feedback" . You may have a look to them. You'll learn that it's an inadequatly named technique and that nothing can make the mathematical difference between between a voltage and a current.

Dartzeel : component of the year ? Ever seen the schematics ?
Probably the worst I ever saw. Overwhelming.
 
#7
forr said:
There already were some discussions at DiyAudio about the so called "current feedback" . You may have a look to them. You'll learn that it's an inadequatly named technique and that nothing can make the mathematical difference between between a voltage and a current.
Click to expand...

I would respectfully disagree. Don't look at DIY discussions

(like here: http://www.diyaudio.com/forums/showthread.php?s=&threadid=36563&highlight= ),

but better at Kester's, Jung's and Mancini's papers, articles and books.
 
#9
ash_dac said:

CFA's disadvantages:-

-poor precision
Click to expand...

What do you mean? Offset? Distortion? I don't think (mis-called) cfb has any inborn limitation to distortion figures. It can be higher or lower than typical vfb. Offset can be higher though.
-poor common mode rejection (due to mismatched input impedances)
Click to expand...
It does not very much matter in typical audio applications.

-poor powersupply rejection
Click to expand...

I don't think so. It rather depends.


-difficult to apply (stray capacitance..)
Click to expand...

I don't understand. Can you say something more?


-high slew rate
-wide bandwidth
Click to expand...

Chose of differential stage being 'voltage' or 'current' type does not limit slew and bandwidth of audio circuits. You can get more than enough using both. For video, measurements, yes, cfb op-amps can have better slew rate (several kV/us) and ultra low settling times.

I use a version of current feedback of low feedback factor and am satisfied with results, both from engineer and listener point of view.

best regards
Adam
 
#15
Current-feedback amp:

Look at the ADI (formerly PMI formerly SSM) ap-note by Mark Alexander on a current-feedback amp. There are two versions, one with Toshiba IGBTs for output transistors and one with power MOSFETS. Otherwise, they are essentially identical. Skip the idea of IGBTs -- the particular devices are history and IGBTS are eminently UNsuitable for audio.

The Alexander design solves the "precision" problem with a very novel servo. Specs are pretty first rate. I don't know about "sound" as I've never heard it, but it's on my list of topologies to play with when I get the time.

Try it. You might like it. Or then, you might not. But be careful about drawing any conclusions about goodness based on topology or a particular implementation of a topology. I've heard a number of people who are very happy with the B-B INA103 as a phone preamp or mic preamp -- it has a CFB input stage.

As this thread demonstrates, the audio world is full of many people with opinions.
 
#20
Re: Re: Current feedback - not suitable for audio ?

darkfenriz said:

-difficult to apply (stray capacitance..)

I don't understand. Can you say something more?

I use a version of current feedback of low feedback factor and am satisfied with results, both from engineer and listener point of view.

best regards
Adam
Click to expand...

http://www.intersil.com/data/wp/wp0588.pdf

http://focus.ti.com/lit/an/slod006b/slod006b.pdf

I read chapter 8, and 9 of op-amps for everyone.

Ok so a dc servo is added for precision in power amplifier designs. I guess precision can be taken off the list. 🙂

This only leaves the layout issue to overcome in a current feedback power amplifier. As others have stated this requires a competent engineer.

How does your CFB amplifier sound ? (I would expect fantastic high frequency performance 🙂 )

Is soundstage related to precision ?
 
Status
Not open for further replies.
Top Bottom