A 9 watts beats Passlabs, EAR and YBA ?

Thanks for the info. You are a font of knowledge!

I understand propogation delay and the need for compensation in the feedback loop to avoid oscillation. The part I don't get is where you say "The result is low-level odd-order harmonic generation". Why that?

Feedback is not necessarily a yes/no thing. The feedback signal can be limited to low frequencies where the propogation delay is insignificant, while the higher frequencies are "zero feedback".

Unfortunately, an SET will not have the drive of the EAR 890, unless you use a very unusual speaker with it's own set of problems. One must come to know one's own predilections.

Hello, Edlertford and Atmasphere,
Special thanks to Atmasphere for given us such detailed explanation on the distortion.
Somehow, choosing the right speakers is the most important thing and in this case; I slept with Maggie.
Maggie is my life now yet SET is my destiny.
You can say that we want to have the cake and eat that too; isn't that every man wants to do that way!
Let's get serious now, someone suggests we try the latest Pass Labs XA series which itself a single-end, class A; Triode or not, I don't know.
All I know trying them is expensive; I have already paid for the X150.
Know anyone who want to buy the X150, brand new, 10 hours of listening.
May be I should give the X150 hours to burn it, right?
See how the bugs bite now.
Help?
The following link certainly describe exactly what we are:

http://www.audionote.co.uk/articles/art_audio_hell.shtml

Go figure, bye
Anyone for more information on helping me?
May be a woman can solve the problem, or even more problems?

Hi, Eldartford, the idea that you can run zero feedback at some audio frequencies and then use feedback at low frequencies suggests to me a gain non-linearity that changes with frequency. So that would take some tinkering to make that work, to say the least. To my knowledge no-one does this, as the idea behind many amps that use feedback is the concept of 'constant voltage' wherein the amp is capable of constant voltage regardless of the load impedance.

Zero feedback amps do not behave this way for the most part, usually subscribing to a different idea wherein the amp makes 'constant power' with respect to the load.

These two ideas have been in conflict in audio for the last 50-60 years, resulting in the tube/transistor debate and the subjectivist/objectivist debate, plus the general idea of component matching.

Robertwolfee, the Pass Labs amplifier is class A, but it is not tube. However it is one of the best-sounding transistor amplifiers out there. At any rate you will need some power for the Maggies, so due to the power limitations of most SETs, you won't get them to be very practical with your speakers although I am sure they they would sound fine, just at very low volume levels.

Atmasphere...True... if the feedback signal had gain increasing with frequency reduction the overall amp would exhibit LF roll off. But that would be easily corrected by equalization of the input signal.

But the causial relationship between propagation delay and odd harmonic distortion still eludes me.