I have seen multiple articles here and elsewhere about the "tube sound."
Some of the theories range from being the 2nd order harmonics to some complex, undefinable distortion given by the Gestalt of the entire amplifier (power, preamp, tube finals, output transformer, etc.)
I have found many mathematical analyses of tubes and distortion effects of a single tube but has anyone narrowed the "tube sound" down to a mathematical model?
If so, has this been verified against a real tube amplifier?
Chris
Some of the theories range from being the 2nd order harmonics to some complex, undefinable distortion given by the Gestalt of the entire amplifier (power, preamp, tube finals, output transformer, etc.)
I have found many mathematical analyses of tubes and distortion effects of a single tube but has anyone narrowed the "tube sound" down to a mathematical model?
If so, has this been verified against a real tube amplifier?
Chris
I would tend to agree with the transformers been a significant component, especially their bandwidth limiting qualities.
Shoog
Shoog
IMHO tube sound
Hi all. my day job is the lead designer of ct scanners front end electrnoics...and my previous job was at MIT as the lead diagnostics engineer for the lab of nuclear science. My hobby is instrumentation, music, and tube hifi. i wrote my MSEE masters thesis on amplifier distortion. It a fun read....some of it is tongue in cheek. http://www.dancheever.com/main/cheever_thesis_final.pdf
what i have learned since then ....building at least another 40 amplifiers...seems obvious...that when i build an amp to sounds good with ZERO feedback, and i apply a little feedback to control the output impedence...this amp will sound MUCH better WAY BEFORE clipping than a ultra high gain amp thats sounds horrible open loop.
its non newtonian in the sense that both amps at, say 1watt, show clean output on a fft...no high order harmonics. but thats with a fixed sine and the fft needs 1000's of cycles to make a picture. So maybe with broad frequency dynamic signals "stuff" interacts in the feedback...stuff that one cant instrument and catch in real time.
Hi all. my day job is the lead designer of ct scanners front end electrnoics...and my previous job was at MIT as the lead diagnostics engineer for the lab of nuclear science. My hobby is instrumentation, music, and tube hifi. i wrote my MSEE masters thesis on amplifier distortion. It a fun read....some of it is tongue in cheek. http://www.dancheever.com/main/cheever_thesis_final.pdf
what i have learned since then ....building at least another 40 amplifiers...seems obvious...that when i build an amp to sounds good with ZERO feedback, and i apply a little feedback to control the output impedence...this amp will sound MUCH better WAY BEFORE clipping than a ultra high gain amp thats sounds horrible open loop.
its non newtonian in the sense that both amps at, say 1watt, show clean output on a fft...no high order harmonics. but thats with a fixed sine and the fft needs 1000's of cycles to make a picture. So maybe with broad frequency dynamic signals "stuff" interacts in the feedback...stuff that one cant instrument and catch in real time.
It depends on what you mean by 'tube sound'. It could mean:
1. nothing at all
2. significant levels of low-order distortion caused by poor bias choice (e.g. cheap Chinese tube buffer) or absence of feedback
3. absence of solid-state-specific distortions
4. other things, perhaps including what others in this thread have said
There are at least two quite distinct groups of people who like valve amplifiers:
A. those who desire 'tube sound'
B. those who wish to avoid 'tube sound'
Solid-state fans tend to assume that A and B are the same thing.
1. nothing at all
2. significant levels of low-order distortion caused by poor bias choice (e.g. cheap Chinese tube buffer) or absence of feedback
3. absence of solid-state-specific distortions
4. other things, perhaps including what others in this thread have said
There are at least two quite distinct groups of people who like valve amplifiers:
A. those who desire 'tube sound'
B. those who wish to avoid 'tube sound'
Solid-state fans tend to assume that A and B are the same thing.
In this article, a well-designed OTL tube amp (Atma-Sphere S-30 Mk.III) is compared to our Mr. Pass's J2. Although the J2 won by a nose, the similarities were greater than the differences:
SoundStageXtreme.com | UltraAudio.com | First Watt J2 Stereo Amplifier
SoundStageXtreme.com | UltraAudio.com | First Watt J2 Stereo Amplifier
dancheever - thanks for the link to your paper. I need some time to read this.
DF96 - I've been thinking about this and it seems like the ideal audio reproduction system would sound exactly like being at a live event. It seems like the "tube sound" that many are trying to reproduce is from memories of previously audio reproductions and not a true reproduction of a performance.
Ideally, I would like to be able to define what "tube sound" is and be able to deconstruct it into the components like 2nd order harmonic distortion, transformer effects, other non-linear effects, etc. I should then be able to (theoretically) model these components and reproduce the tube sound in a simulation that matches a real amplifier.
I have used the ARTA software to look at THD+N of a hybrid tube amplifier (tube preamp + push-pull final) that I built and have the attached response. I have assumed that the 2nd order harmonics are the "tube sound" since when I bypass the tube, it goes away.
If the effects of the magnetics are significant to the system, I would like to figure out what they are. Also, it seems like a transformer output vs a direct output from a push-pull SS amplifier would be much more dependent on the speaker chosen and it's impedance vs frequency.
Chris
DF96 - I've been thinking about this and it seems like the ideal audio reproduction system would sound exactly like being at a live event. It seems like the "tube sound" that many are trying to reproduce is from memories of previously audio reproductions and not a true reproduction of a performance.
Ideally, I would like to be able to define what "tube sound" is and be able to deconstruct it into the components like 2nd order harmonic distortion, transformer effects, other non-linear effects, etc. I should then be able to (theoretically) model these components and reproduce the tube sound in a simulation that matches a real amplifier.
I have used the ARTA software to look at THD+N of a hybrid tube amplifier (tube preamp + push-pull final) that I built and have the attached response. I have assumed that the 2nd order harmonics are the "tube sound" since when I bypass the tube, it goes away.
If the effects of the magnetics are significant to the system, I would like to figure out what they are. Also, it seems like a transformer output vs a direct output from a push-pull SS amplifier would be much more dependent on the speaker chosen and it's impedance vs frequency.
Chris
Attachments
Are you talking about the theory or audible aspect of it? I suspect that talking about things that are audible would be more fun on audio forum.
What do you mean by real? Are you saying there are fake tube amplifiers even though the tubes are used?
My brother-in-law was sold a Chinese tube amp and he never liked the thing. When he gave it to me (yes, it's that Dynaclone that now only needs a EUR200 fix of decent OPTs) it became apparent that the design was seriously flawed. Nor the Chinese manufacturer nor the Dutch tube amp shop understood basic circuit design (preamp grid current! Aaaagh!). Now I've set my goal to have it sound like my solid state amp in an A/B test. So far that hasn't happened. But I just like the use of half-century old technology to get good results. My only concession being semiconductor rectifiers.
BTW- is it me or do the people advocating "tube sound" never use an A/B test setup?
@DF96: guess I'm in the "B" category... 😀
BTW- is it me or do the people advocating "tube sound" never use an A/B test setup?
@DF96: guess I'm in the "B" category... 😀
Not sure what the tube sound is, but I do know what I like and what I hear and it is subjective to the ears of each listener. To me music just sounds smoother through a nicely built SE KT-88 amp.
Chip amps that I have heard sound a little cold and harsh/attack heavy to me (through cheapo speakers, probably no problem, but I would chose my old Pioneer SX-780 with good speakers over chip amps).
And Class A, that sounds just different and exact/sharp, but very good also.
Plus, is there anything more exotic looking and appealing than the warm glow of tubes putting out nice jazz?
Chip amps that I have heard sound a little cold and harsh/attack heavy to me (through cheapo speakers, probably no problem, but I would chose my old Pioneer SX-780 with good speakers over chip amps).
And Class A, that sounds just different and exact/sharp, but very good also.
Plus, is there anything more exotic looking and appealing than the warm glow of tubes putting out nice jazz?
Evenharmonics - I am talking about specifically what distortion / non-linearities that the tube amplifiers add. By "real tube amplifier," I mean a physical piece of hardware and not a simulated (SPICE or other) circuit. I am referring to a tube amplifier that uses high-voltage for the tubes to be operated as intended and not the low-voltage preamps that claim to give tube sound.
If the characteristics of "tube sound" are defined mathematically, one could possibly devise a solid-state (analog or DSP) circuit to simulate the sound. I know that some products simulate the "tube sound" but it isn't clear what they are doing. Did they just to an impulse response on a tube amp and use that for the effect or something else?
Chris
If the characteristics of "tube sound" are defined mathematically, one could possibly devise a solid-state (analog or DSP) circuit to simulate the sound. I know that some products simulate the "tube sound" but it isn't clear what they are doing. Did they just to an impulse response on a tube amp and use that for the effect or something else?
Chris
To point out the obvious elephant in the corner of the room, "tube sound" is quite definite in the overdriven distortion zone favored by electric guitar players and Leslie speaker users (keyboards). Tubes have a gradual 2nd harmonic distortion curve when overdriven which is pleasing enough to these instrumentalists that the tube plants have been kept alive for us hifi people to buy their products. Nothing comes close to this distortion curve, not even a zener clipping circuit.
For hifi use, I view "just like the live concert" to be my standard. My tube amp (ST70) adds a little fuzziness to top octave piano and a little buzz to tinkly bells, that I could do without. My ST120 with djoffe bias mod and the CS800s don't have this noise.
However, one main advantage of tubes is that there is no way to drive them into hard clipping, which is a very annoying sound when produced by a solid state amp that is overdriven. As music tends to be louder at the beginning of the note, perhaps this is what overtheairbroadcast is complaining about with chip amps. Perhaps they are power deficient at the beginning of the note. Even my ST120 has such a high power supply that it can put out 200 w/ch for milliseconds when the music demands it, instead of the 60 w/channel the heat sinks limit it to for rating purposes.
Beyond that, I don't know. I'm not that fond of the slight fuzziness of my tube amp, I just put up with it for 3 decades, and am back to it now that the solid state amps all have problems again. The H182 organ with 23 1968 production tubes and 71 new electrolytic capacitors is in full health, also, making some amazing music while I play after dinner instead of fiddle with electronic boxes.
For hifi use, I view "just like the live concert" to be my standard. My tube amp (ST70) adds a little fuzziness to top octave piano and a little buzz to tinkly bells, that I could do without. My ST120 with djoffe bias mod and the CS800s don't have this noise.
However, one main advantage of tubes is that there is no way to drive them into hard clipping, which is a very annoying sound when produced by a solid state amp that is overdriven. As music tends to be louder at the beginning of the note, perhaps this is what overtheairbroadcast is complaining about with chip amps. Perhaps they are power deficient at the beginning of the note. Even my ST120 has such a high power supply that it can put out 200 w/ch for milliseconds when the music demands it, instead of the 60 w/channel the heat sinks limit it to for rating purposes.
Beyond that, I don't know. I'm not that fond of the slight fuzziness of my tube amp, I just put up with it for 3 decades, and am back to it now that the solid state amps all have problems again. The H182 organ with 23 1968 production tubes and 71 new electrolytic capacitors is in full health, also, making some amazing music while I play after dinner instead of fiddle with electronic boxes.
Last edited:
Along the lines of feedback, there is an interesting discussion here :
Negative feedback in audio amplifiers: Why there is no such thing as too much | EDN
Negative feedback in audio amplifiers: Why there is no such thing as too much | EDN
Well,
I thought tube sound was about the output transformer until I built an OTL.
There is something about the OTL sound that is very tube like "BUT" its not warm or third harmonic, its got more bass drive than most Tx output transformer types.
So what is tube sound? I don't believe its OP transformer sound. Although the TX has an impact.
Yes it is about the illusion of a live performance. Then again isn't all HIFI.
Lack of emotion or non involvement as its sometimes called is non HIFI. Then again "to some" if it measures great then its HIFI even if non involving.
So perhaps its what you want it to be. If you are a violin player and you think music is about emotional involvement then what is HIFI?
Is tube sound HIFI?
Regards
M. Gregg
I thought tube sound was about the output transformer until I built an OTL.
There is something about the OTL sound that is very tube like "BUT" its not warm or third harmonic, its got more bass drive than most Tx output transformer types.
So what is tube sound? I don't believe its OP transformer sound. Although the TX has an impact.
Yes it is about the illusion of a live performance. Then again isn't all HIFI.
Lack of emotion or non involvement as its sometimes called is non HIFI. Then again "to some" if it measures great then its HIFI even if non involving.
So perhaps its what you want it to be. If you are a violin player and you think music is about emotional involvement then what is HIFI?
Is tube sound HIFI?
Regards
M. Gregg
Last edited:
Regards the question,
Can it be put in an equation, yes and no.
Yes if it reproduces the sound exactly as in the studio.
This is, it measures correctly low distortion etc.
_______________________________________________
No there are to many variables like room speaker types etc.
So its the synergy of everything.
Tube sound can sound horrible and so can SS amplification it depends on the system and the room + the listener.
So there you have it, is it an Audiophile or Engineers system.
It can be both however that would be something special (HIFI)
Regards
M. Gregg
Can it be put in an equation, yes and no.
Yes if it reproduces the sound exactly as in the studio.
This is, it measures correctly low distortion etc.
_______________________________________________
No there are to many variables like room speaker types etc.
So its the synergy of everything.
Tube sound can sound horrible and so can SS amplification it depends on the system and the room + the listener.
So there you have it, is it an Audiophile or Engineers system.
It can be both however that would be something special (HIFI)
Regards
M. Gregg
The "tube sound" is bright and forward. It's mellow and laid back. It's startlingly clear. It's liquidly hazy.
It's nonsense.
Tubes have no sound. Amplifiers coupled to speakers have sound. That sound will depend on frequency response, noise, distortion, source impedance, stability, sensitivity to vibration, and overload recovery. It will depend on those things irrespective of the particular choice of parts used to design for these properties.
It's nonsense.
Tubes have no sound. Amplifiers coupled to speakers have sound. That sound will depend on frequency response, noise, distortion, source impedance, stability, sensitivity to vibration, and overload recovery. It will depend on those things irrespective of the particular choice of parts used to design for these properties.
For many years (and maybe still?), an audio engineer called Richard Clark ran an "Amplifier Challenge," where he offered a prize of $10,000 to anyone who could demonstrate in double-blind testing that they could reliably discriminate between any two amplifiers, subject to certain fairly basic conditions. The "challenging" amplifier and the comparison had both to have less than 2% THD at the levels at which they were operated; the signal levels would be carefully matched to 0.05dB accuracy; and if either had a significant deviation from flat frequency response in the audio range, then an equaliser would be used on one or other amplifier (challenger's choice) to match them. Nobody ever won the prize, although of order 1000 have tried.
Typically, as I understand it, if the challenging amplifier were a tube amplifier then Clark would make a little R/C equaliser to match the solid-state comparison amplifier's frequency response to the tube amplifier, and maybe add a series resistor at the output of the SS amplifier to match the output impedance of the tube amplifier. So effectively, by this means, he would mock up the performance of the tube amplifier at the cost of about $5 worth of components, he reckoned.
Different people will place different interpretations on what this shows, I'm sure. To me, it would seem that if no one can hear the difference between the real tube amplifier and the mocked-up one in blind listening tests, then the sounds are the same.
Chris
More details here: Amplifier Challenge Rules
- Status
- Not open for further replies.