In this thread https://hackaday.com/2023/04/04/reshoring-vacuum-tube-manufacturing-one-tube-at-a-time/#respond ,Winston ascribes the "(second-order harmonic distortion from the tube interacting with the inductance of the output transformer)" as the "tube sound" which some find pleasing.
I’m no expert on tube or solid state amp design, but is that described distortion interaction the only likely “euphonic” effect that attracts some people to tube amps; perhaps 300B SET amps, in particular?
At the same time, could the attraction also be largely based on the fact that single-ended tube and solid state amps must all be Class A biased (which otherwise would produce waveform clipping)-and that, perhaps also due little need for negative feedback, might thereby produce a similarly pleasing “warm” sound?
I’m no expert on tube or solid state amp design, but is that described distortion interaction the only likely “euphonic” effect that attracts some people to tube amps; perhaps 300B SET amps, in particular?
At the same time, could the attraction also be largely based on the fact that single-ended tube and solid state amps must all be Class A biased (which otherwise would produce waveform clipping)-and that, perhaps also due little need for negative feedback, might thereby produce a similarly pleasing “warm” sound?
It's a can of worms, you'll likely get as many opinions as posts. So I'll jump in with my experience and opinions. 
IME what a lot of people call the "warm" sound of tubes isn't coloration at all, but a lack of the normal coloration we hear in Class-AB transistor amps. A coloration that many of us have now gotten so used to, that it seems like the default. But tubes amps have their own problems and not all are smooth or warm. Read some reviews from the golden age of Hi-Fi and many tube amps might be called strident or harsh. I've owned some of those.
The 300B can be euphoric if designed that way, but the first I ever heard was Jean Hiraga's 300B that was anything but colored, it was shockingly transparent and neutral. It's a little different in its design than the current crop of 300B or other SET amps, some of which are very warm colored. I remember a particular 45 SET amp that was super warm, but damn did it sound good on Sinatra. Jean Hiraga had a 300B push-pull that was stronger than his SET, fuller and warmer. Nice, but we liked the SET a little better than the P-P, because the the former's transparency.
There are some great Class-A transistor amps that are neutral, clean and smoother than many P-P tubes amps I've heard. That can equal "warm" to some ears, as the normal AB transistor grunge is not there. Good luck with your quest, I know there will be many opinions and conflicting views.
IME what a lot of people call the "warm" sound of tubes isn't coloration at all, but a lack of the normal coloration we hear in Class-AB transistor amps. A coloration that many of us have now gotten so used to, that it seems like the default. But tubes amps have their own problems and not all are smooth or warm. Read some reviews from the golden age of Hi-Fi and many tube amps might be called strident or harsh. I've owned some of those.
The 300B can be euphoric if designed that way, but the first I ever heard was Jean Hiraga's 300B that was anything but colored, it was shockingly transparent and neutral. It's a little different in its design than the current crop of 300B or other SET amps, some of which are very warm colored. I remember a particular 45 SET amp that was super warm, but damn did it sound good on Sinatra.
Jean Hiraga had a 300B push-pull that was stronger than his SET, fuller and warmer. Nice, but we liked the SET a little better than the P-P, because the the former's transparency.There are some great Class-A transistor amps that are neutral, clean and smoother than many P-P tubes amps I've heard. That can equal "warm" to some ears, as the normal AB transistor grunge is not there. Good luck with your quest, I know there will be many opinions and conflicting views.
Could you likewise also reference any Class A solid-state brand/model amps which you may have heard?
Though again, there will undoubtedly also be as many opinions as posts on this begged question, but if a 300B SET amp can, however rarely, be intentionally designed to be neutral and transparent sounding-and a solid-state Class A, if anything, even more so-then how might the "sonic signature" of most Class D power amps (in the ~ $4K range) compare to them? Benchmark? NAD? , et al.
Though again, there will undoubtedly also be as many opinions as posts on this begged question, but if a 300B SET amp can, however rarely, be intentionally designed to be neutral and transparent sounding-and a solid-state Class A, if anything, even more so-then how might the "sonic signature" of most Class D power amps (in the ~ $4K range) compare to them? Benchmark? NAD? , et al.
He's right about the 2HD, which adds sweetness, but what he means by 'interacting with the inductance of the output transformer' is anybody's guess. This is not how transformers work.
Output impedances of a few tenths of an ohm or greater will noticeably shift the speaker crossover points. The usual effect is to increase the overlap between drivers. That may sound subjectively "better", but the results are completely dependent on the speaker.
You can demonstrate this effect by adding 0.22-1.0 ohm series resistors to the outputs of a good amplifier.
Ed
You can demonstrate this effect by adding 0.22-1.0 ohm series resistors to the outputs of a good amplifier.
Ed
Energy is stored in the inductance of the output transformer primary winding. Perhaps for this reason transformers with much higher than the required primary inductance have a more authoritative bass response.
Not sure if dumbing down amplifier sound to a harmonic spectrum and output impedance is very useful. It is by far not so simple. Or so well understood.
I can't speak well to the $4K Class-D amps, I've heard them but don't remember them. The two types that I distinctly remember are Tripath and Icepower, because I spent so much time with them. To me they are more pleasant than most A-B solid state, less artificial. The Icepower very strong and neutral, the Tripath somehow "agreeable" or "pleasing". There has to be some flavor there, because I remember walking down the hall at RMAF one year and hearing music coming out of a room and immediately recognized it as familiar, then as Tripath. I was right. But I don't think I could do that trick today. I sold a fair number of Tripath based amps to people that owned SET amps, and they often liked the Tripath as much or better, so perhaps there was a similar feel to them.
That's the received wisdom on the topic, but it's been hard to prove. Ask forum member @Wavebourn about it, he's done the experiments.
I can say that from my building experience that H3 distortion in the bass can be impressive sounding. It tends to make the bass sound fatter and stronger than it is when clean. A lot of people like it, and maybe it's one of the tricks in the "MegaBass" style enhancements. Go up into the midrange, though, and H3 isn't so nice.
Probably not that simple. If someone listens to a tube amp, takes an FFT and looks for spurs, and if that's all the information you have then what else is there to conclude? You hear a sound, you measure some HD. That must be the whole story as to the cause of the sound, right? I don't think so.
It looks more to me like an illustrative case of WYSIATI bias.
https://www.shortform.com/blog/what-you-see-is-all-there-is/
Yes and no
But for the most part yes.
Rather solid state or tube.
A " Harmonic Profile" with strong even ordered harmonics
or strong Second Harmonic is often considered more pleasing.
Otherwise a can of worms.
Since the phase of the harmonics could be open to interpretation.
And there is plenty of examples where " tube" amps can have very strong
third harmonic or stronger odd harmonics.
The so called horrible things that solid state does.
When countless Solid State designs can have a better
Harmonic profile than a tube amp.
Likewise plenty of " Magical" solid state " Class A " amps
which have very high 3 rd harmonic. Being many are
just overbiased push pull amps.
Basically cooking themselves to death and wasting energy for no
reason. Since the harmonic profile is nothing special.
Often the 3 rd harmonic is even higher with overbiased class A/B
Aka push poo Class A
But for the most part yes.
Rather solid state or tube.
A " Harmonic Profile" with strong even ordered harmonics
or strong Second Harmonic is often considered more pleasing.
Otherwise a can of worms.
Since the phase of the harmonics could be open to interpretation.
And there is plenty of examples where " tube" amps can have very strong
third harmonic or stronger odd harmonics.
The so called horrible things that solid state does.
When countless Solid State designs can have a better
Harmonic profile than a tube amp.
Likewise plenty of " Magical" solid state " Class A " amps
which have very high 3 rd harmonic. Being many are
just overbiased push pull amps.
Basically cooking themselves to death and wasting energy for no
reason. Since the harmonic profile is nothing special.
Often the 3 rd harmonic is even higher with overbiased class A/B
Aka push poo Class A
The fundamental value of a good scientific theory is determined by the extent it can be reliably be used to predict the future.
In the case of theories about HD and amplifier sound, listening to an amplifier and looking at an FFT, then producing an explanation as to causality is flawed.
If HD could predict the sound of an amplifier before anyone listens to it, then the prediction could be tested for causality.
Otherwise, what you are showing is correlation, not causation. To confuse correlation with causation is a fundamental error.
In the case of audio equipment and HD analysis, IIUC HD is a rather weak predictor of SQ. When HD is very bad then it has better predictive value that the sound won't be liked, but when HD is at more normal levels then its predictive power is significantly less reliable.
To frame the problem in another way, Dr. Earl Geddes wrote:
"The bottom line here is that we know so little about how humans perceive the sound quality of an audio system, and in particular the loudspeaker, that one should question almost everything that we think we know about measuring it. From what we have found most of what is being done in this regard is naive.
In the case of theories about HD and amplifier sound, listening to an amplifier and looking at an FFT, then producing an explanation as to causality is flawed.
If HD could predict the sound of an amplifier before anyone listens to it, then the prediction could be tested for causality.
Otherwise, what you are showing is correlation, not causation. To confuse correlation with causation is a fundamental error.
In the case of audio equipment and HD analysis, IIUC HD is a rather weak predictor of SQ. When HD is very bad then it has better predictive value that the sound won't be liked, but when HD is at more normal levels then its predictive power is significantly less reliable.
To frame the problem in another way, Dr. Earl Geddes wrote:
"The bottom line here is that we know so little about how humans perceive the sound quality of an audio system, and in particular the loudspeaker, that one should question almost everything that we think we know about measuring it. From what we have found most of what is being done in this regard is naive.
Last edited:
Adding H3 at low frequency is pretty much like adding some mild clipping. By flat-topping a little bit it sounds louder without requiring increased excursion from the speaker (or a lot more average power). The result is more tactile feel and it sounds like the system is being driven harder.I can say that from my building experience that H3 distortion in the bass can be impressive sounding. It tends to make the bass sound fatter and stronger than it is when clean. A lot of people like it, and maybe it's one of the tricks in the "MegaBass" style enhancements. Go up into the midrange, though, and H3 isn't so nice.
Those “bass enhancers” add 3rd and 5th harmonic in the same proportion you would get if the amp were clipping, without adding the buzzy sounding higher harmonics. Again, it “sounds” like the system is being driven hard and distorting, and the brain “hears” the missing/depressed fundamental that would normally be driving the amp or woofer to distortion.
It works with IM products, too. You know that annoying Amber Alert tone on your phone? You hear that beat tone in your head, even though the speaker in your phone is totally incapable of reproducing it.
Not to mention Direct Radiators being fairly common.
and expect the usual 1 to 3 % distortion at average to higher
listening levels.
Assuming generalized harmonic profile is likely very high
2nd Harmonic. Or even ordered
If the amp doesn't have magic harmonics
your speakers will.
Music industry having developed intentional
distortion devices. I Can agree with designing
symmetrical and asymmetrical clipping devices.
For bass or extreme " crunch" I can say high 3rd harmonic
can give more depth.
Without ways to prove to the world what harmonic profile is
better.
It seems H3 in lower frequency is ok. The classic H2 theory
seems to apply to high frequency being more pleasing.
Or likely more noticeable 2k and above or in known bands where
the ear is more sensitive.
Likewise it is rather easy to design a solid state with "Tube"
second harmonic. But usually above 10k it is very hard to
maintain when the design intentionally creates 2nd harmonic.
Above 10K the 3 rd usually still dominates. Or overall odd ordered
profile at higher frequency.
and expect the usual 1 to 3 % distortion at average to higher
listening levels.
Assuming generalized harmonic profile is likely very high
2nd Harmonic. Or even ordered
If the amp doesn't have magic harmonics
your speakers will.
Music industry having developed intentional
distortion devices. I Can agree with designing
symmetrical and asymmetrical clipping devices.
For bass or extreme " crunch" I can say high 3rd harmonic
can give more depth.
Without ways to prove to the world what harmonic profile is
better.
It seems H3 in lower frequency is ok. The classic H2 theory
seems to apply to high frequency being more pleasing.
Or likely more noticeable 2k and above or in known bands where
the ear is more sensitive.
Likewise it is rather easy to design a solid state with "Tube"
second harmonic. But usually above 10k it is very hard to
maintain when the design intentionally creates 2nd harmonic.
Above 10K the 3 rd usually still dominates. Or overall odd ordered
profile at higher frequency.
Last edited:
Maybe try this for bass sound: https://www.waves.com/plugins/maxxbass
Hey, looks like it does something besides adding harmonics...
There are also tube amp emulators such as: https://www.uaudio.com/uad-plugins/channel-strips/manley-tube-preamp.html
"Add tube detail and rich complex circuit behaviors to make your tracks come alive"
IME the UAD plugins are among the very best or the best. They have some pretty sophisticated DSP emulations of real gear.
One may also notice that the above DSP plugins include some non-time invariant behaviors, much like some real circuits.
Hey, looks like it does something besides adding harmonics...
There are also tube amp emulators such as: https://www.uaudio.com/uad-plugins/channel-strips/manley-tube-preamp.html
"Add tube detail and rich complex circuit behaviors to make your tracks come alive"
IME the UAD plugins are among the very best or the best. They have some pretty sophisticated DSP emulations of real gear.
One may also notice that the above DSP plugins include some non-time invariant behaviors, much like some real circuits.
Last edited:
The vast majority overlook clarity, cleanliness as another important basis for "euphony"-)
The vast majority of "euphonic" amplifiers, mostly attributed to the distortion behavior, are minimal amplifiers: only a few components modulate, noise, dirt the signal with their noise, dirt;-)
And: tube amplifiers are also because of this a super basis for listening euphoria;-)
Avoid highly complex analog complementary transistors push-pull circuits. They rumble gray and flat three tones only;-)
The vast majority of "euphonic" amplifiers, mostly attributed to the distortion behavior, are minimal amplifiers: only a few components modulate, noise, dirt the signal with their noise, dirt;-)
And: tube amplifiers are also because of this a super basis for listening euphoria;-)
Avoid highly complex analog complementary transistors push-pull circuits. They rumble gray and flat three tones only;-)
;-)
May be, these can only counter the sound of bad speakers with their bad sound;-) An old garage punk recording would reveal the "cheating";-)
Yeah, that's a story that has gotten spun out of proportion. He claimed a lot more than the test actually showed. It's a classic audio tale, but bears a heavy dose of scrutiny.