There was an amp that Pavel Macura built, inspired by John Curl's ideas, that used mosfet drivers and BJT outputs, but I am not sure if it was Sziklai configuration. It is possible that it is a kind of IGBT built from discrete transistors, like used in Lazy Cat's TSSA. And Michael Chua from AmpsLab has an VFB amp kit with this combination, but I think it is not Sziklai.
Who can still be stuck as a VFB sectarian member, when so many comments report exactly the same feelings and listening impressions ?
As i tried to say to D.S. i had modified all my mixing desks summing OPA amplifiers from VFB to CFB .
They work in inverting mode in the mixing bus.
You plug only one slice of of the desk, with some nice recorded acoustic guitar as an example. Only this slice in the bus, both OPAs can be very similar.
But, if you plug-in 24 other slices (with no signal in it), the difference begin to be obvious. While the CFA OPA keep the sound unchanged (just little noise added due to the increase of the gain), the sound in the VSA turn dull and agressive.
Now, if you mix 24 instruments, while CFA OPAs keep them just placed one near the other, easy to follow and separate, VFA make a 'real mix' of them, some kind of mashed potatoes, you know, kinda Tamla motown Berry Gordy's wall of sound
Well, there is other interesting tracks to get rid or reduce negative effects of global feedback or servo evils. Like Error correction ( CordellAudio.com - A MOSFET Power Amplifier with Error Correction ) or what i name error cancellation ( http://www.diyaudio.com/forums/solid-state/221309-my-little-posh-tringlophone-6.html#post3196685 ).
Did D.S. talk about those ideas ?
Did D.S. talk about those ideas ?
Last edited:
It is very instructive for circuit designers to follow Esperado's sequence in "research and development". Esperado first noticed that some amps sound better then the others, then he found out that those better sounding amps are CFB, then he tried to understand what makes CFB amps better sounding by using measurements and simulation.
Most circuit designers these days follow exactly the opposite sequence. First simulation, then prototype (perhaps?), then measurements. If simulation and measurements are good circuit designer claims that such amp must be good sounding. The amp could be good sounding. But it ain't necessarily so.
Most circuit designers these days follow exactly the opposite sequence. First simulation, then prototype (perhaps?), then measurements. If simulation and measurements are good circuit designer claims that such amp must be good sounding. The amp could be good sounding. But it ain't necessarily so.
kgrlee,
If I open up my Earthworks mic pre is that what I am going to see? Or is it a more modern version of that circuit with later discrete components?
Esperado,
Now if you could only design a simple cfb amplifier for me with modern devices I would be very happy. Are you saying that the Simple Symmetrical Amplifier is this very same thing?
If I open up my Earthworks mic pre is that what I am going to see? Or is it a more modern version of that circuit with later discrete components?
Esperado,
Now if you could only design a simple cfb amplifier for me with modern devices I would be very happy. Are you saying that the Simple Symmetrical Amplifier is this very same thing?
This type circuit can be done quit well now. There are existing designs or new ones coming into being frequently. It has taken awhile but slowly evolving and being accepted as the best topology for audio.
However, the best direction now will be to use the least number of active devices and combine cancellation techniques with some NFB.
Thx-RNMarsh
This is my current thinking. Although, my thinking grinds to a halt at cancellation techniques. And implementing something like the Baxandal VAS.
Yes, Kindhornman, (nice to meet-you again) as well as the VSSA http://www.diyaudio.com/forums/vendors-bazaar/225747-vssa-lateral-mosfet-amplifier.html
If this last one is powerful enough for you, there is a group buy for a near complete kit (printed board, SMD aboard, complete active devices).
L.C. pretend VSSA to be yet better sounding than original more complicated SSA.
Both are CFA with a double feedback path allowing direct coupling of input as well as feedback, as the two side of the feedback cancel themselves the base/emitter diode voltage drop of the input bipolar transistor.
And L.C. is that kind of guy who design its projects in real world with his (accurate) ears, and tune them with hours of listening
It fulfill the clever (as always) requirement of Richard:
____________________________________________________________________________________________
Names ? (I'm very suspicious about so calling 'Gurus' ;-). I better rely on listening comments of ordinary humans with taste for music.
Thanks for your kind words, ivanlukic. In fact, i began my professional life in audio design at a transition time between tubes and transistor. When i entered in the R&D department, the company witch engaged-me was famous for their cheap poor sounding amps and kits (copy of RCA databook schematics, 2N3055 based). They measured better than tube amps, sounded horrible on my point of view while tube amps were so nice.
I was young and not very experienced in audio design, but what interested my boss was,with a little background as a sound engineer too, i was mostly involved in 'music reproduction'( my mother was a pianist).
Our first work was to design a "good sounding prestige amp". We produced an amp, using the fastest power devices available at this time (planar epitaxial BDY56 ) after intensive studies of the influence of slewrate, settling time, and printed board good practices. It happened that this amp used a current feedback topology. It was just because this amp was better at our ears than various other VFB we tried. But we had no special generalized thoughts about CFB at this time.
I was very choked, talking to the nice Scott Wurcer guru when he said that, in the Analog Design R&D department, they don't have even a decent hifi system to listen to the OPA they designed. While my favorite CFB OPA was from AD.
On our side, we used our expensive measuring instruments (like distortiometers) more to bring nice numbers to the marketing department once prototype where finished, than to drive blind our researches. We listened a lot of music at each step of the development. It was sometime hard to justify this to our boss: "Am i paying-you to listen to music all the day long ?"
Ears are nice measurement instruments, with practice once you know what to listen to, and always plugged.
I believe we are at the same transition time with Class D amplifiers, and follow works like Bruno's ones with a lot of interest. Full of promises as switching active devices are faster and faster.
Last edited:
On the contrary, ears are absolutely the worst thing you could use to "measure" electronic equipment.
michaelkiwanuka,
Haven't we been down this road long ago. All the early Japanese solid state equipment measured almost no distortion with their test protocols and we all agree that the majority sounded terrible. I am not in any way saying to ignore the test equipment, that would be stupid, but the final test always seems to be do people like the sound when everything is finished. If it measures great but sounds terrible who is right the machine or the ears?
Haven't we been down this road long ago. All the early Japanese solid state equipment measured almost no distortion with their test protocols and we all agree that the majority sounded terrible. I am not in any way saying to ignore the test equipment, that would be stupid, but the final test always seems to be do people like the sound when everything is finished. If it measures great but sounds terrible who is right the machine or the ears?
When I hear something I don't like I go looking for it in the measurements . Seldom is it obvious . Also the lesson is often not universal once discovered .
I sometimes use the inverting input for amps . This limits me to 3K input impedance . To be honest I doubt I ever preferred it ( yet ) .
With that in mind my next pre amp will have 8 watts 8R out put . That way it will drive any input and give me a chance to really know . A discrete op amp would be an idea . The Hiraga Le Monstre seems an ideal . It also means my preamp can be used in many places .
I sometimes use the inverting input for amps . This limits me to 3K input impedance . To be honest I doubt I ever preferred it ( yet ) .
With that in mind my next pre amp will have 8 watts 8R out put . That way it will drive any input and give me a chance to really know . A discrete op amp would be an idea . The Hiraga Le Monstre seems an ideal . It also means my preamp can be used in many places .
On the contrary, ears are absolutely the worst thing you could use to "measure" electronic equipment.
But even if the human ear is that unreliable an instrument, it's still the only one that really matters. All others are just tools that help us to try to understand things.
No, we don't. I hope you are not comparing Mark Levinson No. 33 with a Japanese boom box. Sansui, for example, were among the best audio amplifiers ever built. More than half the innovations/patents in audio amplifiers came at the time from Japan audio designers and manufacturers.
It's an American cultural thing to necessary dislike any Japanese audio devices. Most likely, because they were manufactured at the same time with the cars that put a strain on the US auto industry.
Can you imagine someone saying that woman is not right for you as the computer says it . As stupid as it sounds this was done . A university determined that beauty was perceived as health . This means healthy children . Rules were drawn up and symmetry was mostly what it was about . Marylin Munroe did not pass the test , her eyes were too close to her nose . Before anyone says the obvious I am not especially attracted to women just because of the 38 in the 38 24 38 she might be ( vital statistics as they called them ) . It was her face .
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.