How much innovation is possible with Discrete Class-A Amplifiers?

[dgardner]

With all the passion that's been applied to the development of great discrete class-a amplifiers, I have to wonder how much real innovation is possible?

After reading several of the books and articles from Self, Sloan, Pass, Borbely, Leach, and others, I can't see where someone is clearly able to innovate.

There seems to be a dozen or so well respected ways to design differential input stages and two camps of religion - BJTs and FETs.

There seems to be a limited number of well accepted ways to design a VAS stage. There seems to be lots of subtleties here.

As for the output stage, assuming you limit the discussion for a moment to pure class-a, there are dozens of variations of complimentary push-pull output stages and lots of variations using BJTs and MOSFETs. Plenty of religion here too.

So, at first glance, it seems that new and innovative contributions to "the sport" are likely to be re-arranging the plethora of variations into some unique combination to suit a particular set of design goals. Or am I completely missing something.

I am amazed at the number of class-a mono blocks on the market and the number of DIY class-a design variations in circulation.

Where is the actual cutting edge these days? Super secret custom silicon? Radical gain stage toplogies? Ultra wide bandwidth extremism? Ultra low noise extremism?

Inquiring minds want to know.....

[Charles Hansen]

With all the passion that's been applied to the development of great music, I have to wonder how much real innovation is possible?

After reading several of the books and articles from Beethoven, Bach, Mozart, McCartney, Hendrix, Coltrane, Davis, and others, I can't see where someone is clearly able to innovate.

So, at first glance, it seems that new and innovative contributions to "the sport" are likely to be re-arranging the plethora of variations into some unique combination to suit a particular set of compositions. Or am I completely missing something?

I am amazed at the number of CDs on the market and the number of song variations in circulation.

Where is the actual cutting edge these days? Atonality? Polyrhythm? Multi-harmonic shifts? Telepathic melody transfer?

Inquiring minds want to know.....

[Nelson Pass]

Both posts are well put.

I recommend reading John Horton's "The End of Science" one
of my favorite books.

[Leolabs]

How about PS Audio's Hybrid Class A????Not an innovative design too??

[amplifierguru]

How about my new upcoming commercial 50W Class A - consumes only 110W and is R2R with only 2 voltage gain stages, no Vas as such and all discrete for < 0.005% THD, as soon as IP is bedded down.

Good reading "Design for the Real World" Victor Papanek ( from Purdue University) and designer of the inflatable sex toy doll, on designing things the world needs rather than wants.


Cheers,
Greg

[PRR]

> Or am I completely missing something.

The details.

The topology is a minor point. Many-many ways to skin cats, most are no good, but there are many ways that work and give essentially the same results.

The devil is in the details. How much current flows past each node, the relation of signal to idle current, the balance of DC power against universal capacitance, getting the most out of never-unlimited money+time. Changing one resistor from 200Ω to 2K2, or vice versa, can make a world of difference. Good designers get their heads down inside the wires and walk with the electrons, feel what they feel, and smooth the path.

Or to abuse Charles' abuse of your question: J.S.Bach did the details better (on average) than most of the Bachs who came after him, who used similar musical forms without the same careful detailing that seemed to spill off of Johan's quill. I won't name some other keyboard composers who, using essentially the same twelve notes that JSB used, can't put them together so well.

To answer your question: complementary devices may have been the last real innovation. Having only N-types does rule out some topologies that are sweet for bipolar signals like audio, and can simplify unipolar systems like supply regulators. But as for the differences between tubes (hollow unipolar FETs) and BJTs (positive-grid FETs/tubes), or using more devices in more ways: bah. Just smoke and mirrors. Good techno-babble for marketing.

> re-arranging the plethora of variations into some unique combination to suit a particular set of design goals.

Sure. Bach just wrote the same 12 notes (not even all 12 at once) in different ways. And sometimes our design goals are no loftier than Bach's: he needed a "new" work every wednesday and sunday to keep the paychecks flowing.

> Where is the actual cutting edge these days?

Marketing, Sales, and Production. Can you trick people into buying boxes faster than your supplier bills come due? If you can't, you will go broke. If you can. it hardly matters what is inside the box. (DIY is an exception to the above, but very minor in the overall picture.)

[Upupa Epops]

All these searching of " superamp with supersound " have own limits : digital format of recording. Is funny to listen neverending discussins and comparisions of sound of amps, which all are " loaded " from " tin " with synthetic sound. Gentlemen, what do you need upgrade in amp architecture ? All is now only searching, how to better hide " unnaturality " of recording. Make sometimes comparision between direct signal from console ( but attention, many consoles have horrible sound ) and signal, which comes from ad/da - at his time you will know, whatabout I'm talking.

[destroyer X]

That ancient paper moving is really a shame!

regards,

Carlos

[mikeks]

Originally posted by PRR
> Or am I completely missing something.

The details.

The topology is a minor point. Many-many ways to skin cats, most are no good, but there are many ways that work and give essentially the same results.

The devil is in the details. How much current flows past each node, the relation of signal to idle current, the balance of DC power against universal capacitance, getting the most out of never-unlimited money+time. Changing one resistor from 200© to 2K2, or vice versa, can make a world of difference. Good designers get their heads down inside the wires and walk with the electrons, feel what they feel, and smooth the path.

Or to abuse Charles' abuse of your question: J.S.Bach did the details better (on average) than most of the Bachs who came after him, who used similar musical forms without the same careful detailing that seemed to spill off of Johan's quill. I won't name some other keyboard composers who, using essentially the same twelve notes that JSB used, can't put them together so well.

To answer your question: complementary devices may have been the last real innovation. Having only N-types does rule out some topologies that are sweet for bipolar signals like audio, and can simplify unipolar systems like supply regulators. But as for the differences between tubes (hollow unipolar FETs) and BJTs (positive-grid FETs/tubes), or using more devices in more ways: bah. Just smoke and mirrors. Good techno-babble for marketing.

> re-arranging the plethora of variations into some unique combination to suit a particular set of design goals.

Sure. Bach just wrote the same 12 notes (not even all 12 at once) in different ways. And sometimes our design goals are no loftier than Bach's: he needed a "new" work every wednesday and sunday to keep the paychecks flowing.

> Where is the actual cutting edge these days?

Marketing, Sales, and Production. Can you trick people into buying boxes faster than your supplier bills come due? If you can't, you will go broke. If you can. it hardly matters what is inside the box. (DIY is an exception to the above, but very minor in the overall picture.)

Amen...in all respects...

[tlf9999]

I don't think you will see much "foundamental" / breakthrough type advances. in terms of electronics, it is fairly well understood. In terms of human hearing, it is not.