Is the appeal to euphonic distortion learned?

I will add that my experiences suggest that what one hears from the typical audience perspective  can be quite different than what one hears from the performance band stand.

That's certainly true!

 As a personal aside, as a young man, for several years I attended musical events almost every night of the week, most of which were unamplified. To this day I find that while tube amps can sound pretty, they don’t sound like the live performances I attended...unless a musician used a tube instrument amp during the performance.

I'm guessing that isn't learned other than your ear is a bit more trained and you've not heard the right tube amp... yet... ?


I play bass and keyboards and from junior high through college and afterwards I played the string bass in a variety of orchestras around town, as well as jazz and folk ensembles. You'd think I would have learned the same thing as you but I didn't, I found that solid state amps of the time were simply incapable of a natural presentation. So it was the opposite 'learning' from yourself (and not discounting your experience in any way).


Since both experiences are subjective and anecdotal, is that the sort of thing we can ignore for this conversation or is that exactly what this conversation is about??


 

@mapman FWIW we are beta testing our class D amps right now. If we don't run into any emissions issues we expect they will begin production later this spring.

if the increased ratio of second and/or third order harmonics in tube equipment masks the higher order harmonics, are you also saying that the frequency response of tube amps in the highest audible frequencies is NOT lower than that of SS amps, but we just don’t "hear" those higher frequencies because they are masked by the 2d/3d orders?

Yes.

Because that means nothing at all comparing tubes/versus S.S. in general...

My post was not meant to address tubes/transistors, it was meant to address how distortion interacts with our ears. What I did not mention in that post is that if you can apply enough feedback to a solid state amp it will sound very natural and the typical brightness and harshness on top will be gone. But that is a tricky thing to do because of something called ’phase margin’ in audio circuits. Essentially there are frequency poles which cause phase shift. At some high frequency, these poles cause enough phase shift that feedback becomes positive rather than negative and the amp will oscillate. This is why for many decades amps have not had enough feedback; with insufficient feedback the circuit will be brighter and harsher than real life. Hence the ’solid state’ sound of the last 50 years.

By the way the human hearing faculties are way less more deceptive that "skeptic sunday scientist" says they are... Reason is simple, evolution.... recognizing timbre voice speech and the source of sounds is vital tool for survival.... The first sense to be born is hearing, the baby listen his mother voice, and in coma and death you can speak with people, it is the last sense organ to go....

If you want to talk about evolution, the ear is the most sensitive at birdsong frequencies as the birds are the first warning that a predator is in the area. This is the Fletcher-Munson curve.

The sound was pleasant but not $100k worth to me. Euphonic as all hell. Detail I know is there was gone, replaced with a warm blanket of sound. So is the loss of detail and an abundance of warmth a result of frequency response or an excess of 2nd harmonic distortion?

This sure sounds to me like an amplifier with too little feedback. Distortion of all types will reduce detail. You know you are making progress when you experience smoothness and more detail at the same time. The smoothness is a lack of higher ordered harmonic distortion and IMD; the detail comes from that lack of distortion.


Amps that exhibit a 2nd order as their primary distortion component have what is mathematically a ’quadratic non-linearity’. A good example of this is an SET where the 2nd harmonic dominates. But the higher orders do not fall off all that quickly; it is only the 2nd (and the 3rd which appears at a lower level) that mask the presence of these higher orders. Because the distortion is so profound, quite simply there is detail lost while sounding otherwise quite smooth.

For instance, I love the Conrad Johnson Premiere 12. They are the ne plus ultra of amplifiers in my mind. Neutral? Hardly! But I can’t really explain why. Those are just very colorful seductive amps.

I was wondering about distortion specifically.

When you combine single-ended and push-pull you have two non-linearities- quadratic and cubic. The Cubic expresses the 3rd as its dominant harmonic component. Due to algebraic summing, this type of circuit typically also exhibits a bit more prominent 5th harmonic. But it will have a lot of the 2nd as well. Distortion does not fall off all that quickly as the order of the harmonic is increased. This is the traditional ’tube sound’ coming out of the golden age of hifi with such amps as the Citation 2 or Dynaco ST-70. Feedback is used to control distortion and to a limited extent reduce the output impedance, but 12-15dB is typical as much more than that risks oscillation. So you get a rich sound (not as rich as SETs) and that 5th sort of acts like a touch of detail.


IMO/IME if you want to use a tube amplifier, getting away from that 2nd harmonic will really allow the amp to be far more neutral and transparent. The only way to do this I know of is to run the amp fully differential from input to output, so that it mostly expresses a 3rd harmonic as the dominant. **This is a good practice for solid state too** and for exactly the same reason! The 3rd is treated the same by the ear as the 2nd but in such circuits often occurs about 1/10th the amplitude as the 2nd is in SETs (this is assuming open loop- no feedback). The tricky bit is that even orders are cancelled not just in the output section but in each stage so distortion isn’t compounded from stage to stage through the amp. For this reason amps with a cubic non-linearity have the succeeding harmonic orders falling off at a faster rate (hence ’cubic’). The reduced higher orders allows the circuit to be smoother and with more detail!

I have one special tube amp and class Ds, can’t seem to decide between them.

Class D amps are a special case of solid state. They can be run with zero feedback (the big trick with zero feedback examples is that RFI and EMI emissions are hard to control), a little or really quite a lot. The high feedback amps have feedback so high that they actually are designed to exceed their own phase margins and go into oscillation! This type of class D amp is known as ’self oscillating’; the oscillation is used as the switching frequency. They have the advantage of being able to use prodigious amounts of feedback in excess of 35dB because they are already oscillating. This allows them to clean up the distortion caused by the feedback itself (which tends to be higher ordered harmonics and is why amps with insufficient feedback have a brightness to them that isn’t natural). So such amps can sound very smooth, much like a really good tube amp.

I don't think its learned.

The presence of enough 2nd or 3rd (both are treated by the ear the same way) masks the presence of the higher ordered harmonics. This is common in tube equipment but not so much in solid state equipment, although Nelson Pass (as an example) is careful to see to it that his stuff expresses a bit more of the lower orders than many solid state products to create this masking.

I'm not saying that his stuff makes more distortion. I am saying that the **ratio** of lower orders to higher orders is enough that the lower orders are able to mask the higher orders to a certain extent. So you can still have lower distortion, say 0.05%THD and still get this masking to work.


Traditional solid state has typically been lower distortion but since the higher ordered harmonics aren't masked solid state has gotten a reputation for being bright and harsh, since the ear is keenly sensitive to these harmonics and assigns a tonality to all forms of distortion. This is why tubes are still around.

Its arguable that the orders of the harmonics should have a weighting system applied, since the ear is insensitive to the lower orders and very sensitive to the higher orders.


When the music is correctly reproduced, the higher ordered harmonics won't be audible. But when the 2nd in particular is too profound, the electronics are said to be 'euphonic', which is to say that they are musical, just perhaps not all that neutral.


But this is certainly not learned! All humans respond to these harmonics in the same way; the higher orders are used by the ear to sense sound pressure. That is why the ear is so sensitive to them.