CFA Topology Audio Amplifiers

Jump to Latest
#3,501
Waly said:
Not sure I understand your point about slew rate..
Click to expand...
Although it seems not difficult to understand.
Richard, as many of us, has noticed some aspects he prefer in the quality of sound reproduced by CFAs.
As he his an experienced designer and engineer, and a wise man, he don't believe he knows everything. So, and like many of us, he 'presume' this superiority to be based on the slew rate difference. Because it is the most obvious one. But how can we be sure ?

Remember Richard in the one who, after noticing a difference between passive component's sound, made a very serious measurement campaign to figure out the objective and technical reasons of those differences. Published by AES.
And, now even pure objectivists can't ignore those sound differences and their reasons, and can't laugh any more against those pretending hearing such differences.

We, (this thread's contributors) love CFAs. It has been demonstrated with simulations and measurements that all the legends against CFA were just legends (Distortion, noise, instability*), and that CFA can equal the best VFAs on all the measurements and even win on several points. Both comparing very simple designs and very sophisticated ones, both topoligies.

This was for the measurements 'objective points.
Now, for the subjective ones (we build amplifier to listen to music) and, as an example, go to the VSSA thread, and read the numerous listening reports i can resume like 'I discover details in my favorite records i never heard before...'. 'easy to listen to, and not fatiguing...'. All correlated. Are they all stupid ?

So, instead of continuing your 'believer' insane campaign against CFAs, why don't you build one, and, for the first time, listen to it, in order to begin to have a little idea of what we are talking of in this thread ?
THAT would be a 'scientific' attitude.
I'm sorry to tell you this, but you act exactly like a member of the Holly Office during Galileo's prosecution. Unable to question your blind faith.

*About stability and real bandwidths: http://www.diyaudio.com/forums/vendors-bazaar/225747-vssa-lateral-mosfet-amplifier.html#post3762113
 
Last edited:
#3,505
scott wurcer said:
The second plot in post #3118 is probably oscillating at many MHz at a low level (digital scopes s**k).
Click to expand...
How to know ? If you refer to the little blurr, it is the same on the input signal, so...
Note, for the scale, that the signal is 2.7MHz, reported phase shift: 72°, gain at this frequency -3dB from flat bandwidth.
I believe L.C. has a analog scope too, but don't know its speed.
I face the same kind of mystery with my poor man's analog 10MHz scope 🙂
 
Last edited:
#3,507
I have a Philips 200 MHz analog scope and a Rigol. The Rigol 1 GS/s (IIRC) does a remarkable job of triggering in the presence of lots of phase noise. I was able to pick up a c. 180 MHz cascode problem on my e-Amp that woudl not have been visible with a 10 MHz scope ( I had a Hitachi dating back to '81 - payment for some design work done for a friend).
 
#3,510
scott wurcer said:
The second plot in post #3118 is probably oscillating at many MHz at a low level (digital scopes s**k).
Click to expand...

No oscillations here, same as input same the output. Digital scopes LCD presentation is just like that, the same artefacts on first 50 kHz plot. Also 2,77 MHz sine input signal is not the best one (old Elektor generator at its max) so it is the output signal, acuratelly following the input signal. That's the way I like it hehe
 
#3,511
wrong on the interweb

Esperado said:
Although it seems not difficult to understand.
Richard, as many of us, has noticed some aspects he prefer in the quality of sound reproduced by CFAs.
As he his an experienced designer and engineer, and a wise man, he don't believe he knows everything. So, and like many of us, he 'presume' this superiority to be based on the slew rate difference. Because it is the most obvious one. But how can we be sure ?

Remember Richard in the one who, after noticing a difference between passive component's sound, made a very serious measurement campaign to figure out the objective and technical reasons of those differences. Published by AES.
And, now even pure objectivists can't ignore those sound differences and their reasons, and can't laugh any more against those pretending hearing such differences.
...
Click to expand...

Jung, Marsh "Picking Capacitors" appeared in Audio - a popular magazine, not peer reviewed, not even a "trade" magazine for a technical professional audience - although it did publish some quite technical articles occasionally
http://en.wikipedia.org/wiki/Audio_(magazine)

a quick search for "Marsh" at the AES library doesn't give any articles by Richard, but does show hits where others cited the "Picking Capacitors" articles

WaltJung Classic Articles has the "Picking Capacitors" article
a quick skim finds "a single blind test" talking point - no data, scores, ranking system explaination or controls cited



more recent:

Capacitor "Sound" in Microphone Preamplifier DC Blocking and HPF Applications: Comparing Measurements to Listening Tests
Author:Gaskell, Robert-Eric
Affiliation:McGill University, Montreal, Quebec, Canada
AES Convention:130 (May 2011)Paper Number: 8350
Publication Date:May 13, 2011

The listening test results were inconclusive for both
the HPF and DC blocking applications. This does
not mean that there is no coloration of the sound
from HPFs or DC blocking capacitors, only that the
effect is subtle enough to be hard to pin point in
laboratory listening tests.
Click to expand...
 
Last edited:
#3,513
jcx said:
Jung, Marsh "Picking Capacitors" appeared in Audio - a popular magazine, not peer reviewed, not even a "trade" magazine for a technical professional audience - although it did publish some quite technical articles occasionally
http://en.wikipedia.org/wiki/Audio_(magazine)

WaltJung Classic Articles has the "Picking Capacitors" article
a quick skim finds "a single blind test" talking point - no data, scores, ranking system explaination or controls cited
Click to expand...

Comment -- the ranking details were not part of the AUDIO article -- a DIY magazine -- that would have been another article in length. Members of the AES (S.Lips et al) did not want it published in the AES. Walt got a lot of flack from the establishment on this subject. he was severely chastised by academics and it was impacting his good reputation. It became political. Even recently such publications as WW and D.Self's own tests focus only/mostly on the coupling cap as distortion sources.

Recall at the time coupling caps were electrolytics in SS equipment. [Tube circuits used mylar.... a better cap dielectric.] Which directly led me to develop direct coupled designs and servo control concept to eliminate the sound (and cost) of coupling cap.

It has still been all but ignore filter apps by established authors and to just talk about coupling cap app vs distortion.

The point was a is that in ANY application, such as filters (passive or active), when a voltage drop is across the cap, the dielectric will an be important source of distortion. (Filters, such as found in RIAA, crossovers , LP, HP filters, EQ, Compensation, etc) And, in these apps it is easily measured dielectric affect. So, in the long, long run we were and are correct about this subject. But, then we already knew that. 🙂

Thx-RNMarsh

Foot-note: It was DIY'ers who tried different cap dielectrics and first heard a difference that lead me to do the research to determine Why or How a cap could make any audible difference. There is more to say about this which hasn't been written about... but that's for another time or life?
 
Last edited:
#3,514
cross post/my addendum

more recent:

Capacitor "Sound" in Microphone Preamplifier DC Blocking and HPF Applications: Comparing Measurements to Listening Tests
Author:Gaskell, Robert-Eric
Affiliation:McGill University, Montreal, Quebec, Canada
AES Convention:130 (May 2011)Paper Number: 8350
Publication Date:May 13, 2011

The listening test results were inconclusive for both
the HPF and DC blocking applications. This does
not mean that there is no coloration of the sound
from HPFs or DC blocking capacitors, only that the
effect is subtle enough to be hard to pin point in
laboratory listening tests.
Click to expand...
 
#3,515
Esperado said:
Lord, we were supposed to design for audio frequencies !
Now how can-you be sure that you have no problems at GHz ?:whip: ;-)
Click to expand...

Just for my audio development only, of course, I use my HP8753D, Anritsu MS420B, Instek GSP810 and a TEK 7000 and TEK TDS3032 (w/fft opt) scope. Nothing less will tell you if everything is OK to 6Ghz.

🙄 :xeye::gasp:

-RM

- digital scopes s**k big time.
 
Last edited:
#3,516
jcx said:
more recent:

Capacitor "Sound" in Microphone Preamplifier DC Blocking and HPF Applications: Comparing Measurements to Listening Tests
Author:Gaskell, Robert-Eric
Affiliation:McGill University, Montreal, Quebec, Canada
AES Convention:130 (May 2011)Paper Number: 8350
Publication Date:May 13, 2011
Click to expand...

Just a wild guess -- wasn't any audio freq signal voltage dropped across the cap?

-RM
 
#3,518
Bigun said:
One way to look at it - most capacitors have better linearity than most inductors, transistors, FETs, tubes, and speakers. If you can, stick to good resistors 😀
Click to expand...

Speaking of good resistors .... a CFA needs good ones for the FB return.

Bob C's comment on the thermal issue has me worried 😕 .

http://www.irctt.com/file.aspx?product_id=36&file_type=datasheet

Would the 3W device above do the job ? My CFA's dissipate a little less current in
the FB network (1+W).
Are there any differences between thin/thick film ?
PS - A small drawback for CFA - special resistors ?? 😀
OS
 
You must log in or register to reply here.
Top Bottom