What is the "Tube Sound"?

[SY]

Yet the question from the OP was about a "tube sound" which could be modelled mathematically, with a desire to model what people perceive as a "tube sound" whether based on bad designs on not.

Essentially, all this talk of there not being a tube sound or ss sound if the amps are designed properly is quite frankly off-topic!

Before you can mathematically model something, you have to show that it exists and be able to quantify it. Some who are mathematically adept and have long experience with tube amps are trying to see if there's anything beyond the usual mundane and extremely well-understood basic engineering parameters- most notably frequency response and the related source impedance as well as distortion. So indeed, it's on topic except for the distraction of trying to refocus things on actual sound and away from handwaving and legend.

There's no mathematical model for fairies.

 

[nigelwright7557]

Someone designed a module to reproduce the "overdriven valve sound" in the 1980's.
I built one and have used it with an electric guitar for many years.
It has a pot to adjust how much distortion is produced.

An externally hosted image should be here but it was not working when we last tested it.

 

[indianajo]

I'm with somebody that said mediocre tube amps have a particular sound and mediocre transistor amps have another two (mediocre transistor amps have hard clipping when loud and crossover distortion when too soft). High end products are different I am sure and not so typecast by technology. I'm also agree with the opinion of fas42, piano sound is so hard to do it is never totally right, only I don't use fuzzy terms like gesault to discuss the things wrong. To be specific my mid-fi ST70 tube amp adds fast vibrato (pitch instability) to top octave piano notes, when pianos have no vibrato.
Per post #130 where Sy mentioned JanFest October 2013, Bing search engine is sure I mean JamFest the cheer and dance competition. Nothing is listed under events and clubs on diyaudio.com, and if this event is not as private as ones at a person's house, I'd like to fantasize about attending: To hear some real high end audio equipment, without buying it. Sy is too popular for PM's I see. Thanks.

 

[Pano]

Send SY an email - he prefers it. And yes, JanFest is held chez SY, wherever that happens to be at the moment.

 

[Johan Potgieter]

May I firstly say that this thread contains some of the best arguments I have encountered in a long span of folks 'unable to keep to the road', although NATDBERG feels that that is what has happened here. (So often, the narrower the 'bandwidth' of argument, the more unanswered questions remain, leading to repetition of subjects. E.g. the case of cables ... OOPS - SORRY! 😱)

...... a circuit that does the least amount of harm - that should be the very basis of the art. But many people simply don't want that, they want enhanced reality. And while we may object to that on some philosophical grounds - are people who want that really wrong?

That is the very basis of my previous point - and also including the similar points made by AndyJevans. And when this is realised, that is where taste comes in and the job of the engineer in a way ends - or at least becomes another job. In a way the definition of "high fidelity" saves the designer!

Not that there is anything wrong with the addition of enhancement; this point was elegantly illustrated by Pano and Andy! But unless we treat those things separately there will always be dissentment; the objectives are simply different and often contradictory. With enhancement the amplifier becomes, in whatever small measure, a musical instrument. With guitar amplifiers this is most noticeable - but that is what the guitarist wants; his amplifier is part of his 'sound footprint' if I can use that term. As I have said in the past: When producing music the guitarist wants whatever he desires to give him a certain characteristic sound. When playing back (reproducing) a recording of his music, he wants a high fidelity amplifier.

Pano, what you (and Andy) illustrate is perhaps the difference between a high fidelity and a high niceness amplifier. (Hi-fi vs.Hi-ni?) And that again is separate from the 'characteristic' sound of mediocre amplifers. Indeed, the field is alive with pitfalls .....

 

[Johan Potgieter]

I have to wonder how all our engineering gurus would fair in comparing real world amps. I suspect not that well.

You have me confused here, Shoog. I thought that 'real world' amps were also designed by 'engineering gurus'. (What exactly do you mean by 'real world' amps?)

 

[DF96]

Why is one person's distortion inferior to another person's fidelity?

Anyone asking a silly question like that should not expect a serious answer.

This thread would not exist if everyone was satisfied with the results of DBT. DBT tells us many things, as Sy can relate in detail. It tells us the elements we can reliably distinguish and those we can't. It can give us good information about how these elements relate to what we can measure, just as Sy has said. We have a lot of information here. What I and others find interesting is why listeners express preferences.

People may prefer all sorts of things. DBT can tell us little about that, except that if preferences disappear when ears alone are used then the preference is not based on sound but some other parameter (e.g. price, fashion, designer's guru status).

Why would we equalize the amps (with all the associated effects) before we would allow a comparison ?

All very weird logic at play here. It seems the engineering fundamentalists want to impose their biases on the consumer.

Not at all. This is simply an exercise in removing differences which are not intrinsic to tubes or SS yet are known to be easy to distinguish by ear alone. Without this the comparison is not between tubes and SS but between one frequency response and another. When testing it is important to ensure that you are testing what you say you are testing and not something different. Non-engineering fundamentalists don't always grasp this so their tests can sometimes be meaningless (or, at least, mislabelled).

 

[Johan Potgieter]

As an aside, pardon off-topic technicalities, just to illustrate some misconceptions/urban legends and while the points have been raised. (No desire to hi-jack thread):

I find it a somewhat odd notion that before you would compare a valve amp to a SS amp you would seek to take away many of the qualities which are fundamental to valve amps.........Why would we equalize the amps (with all the associated effects) before we would allow a comparison ?

Quite. That sort of equalising should not happen when comparing.

Is not lower damping factors a quality of almost all valve amps?.

No, mostly inferior ones, or as said before, where a degree of 'niceness' is purposely built in - for those who desire such. Valve amplifiers can (should) have damping factors of >30. I have built one with a DF of >100 (and I am not talking about motional feedback).

Isn't a tailed off frequency response almost universal to valve amp implementations using real world OT?

No. Decent valve amplifiers have an open-loop frequency response of at least 20kHz. (The classic Williamson has a closed-loop frequency response of 2Hz - 200kHz.) Good OPTs can go up to >50kHz. And while on the subject:

A.wayne: No, any proper OPT should have no specific 'sonic signature'. It is fairly easy to design an OPT with distortion lower than the output valves and a frequency response of 20Hz to 50kHz.

Aren't these part of the package when someone goes looking for the valve sound, shouldn't these qualities be considered on their own terms ??

If so, that should not necessarily be. Sure, as said before, a client has a right to whatever he wants. But it should not occur by 'default', in a manner of speaking, as is sadly the case with many contemporary designs. He would not have found those idiosyncrasies with classic valve amplifiers like Leak, Williamson, Quad II, Radford and the like.

 

[TiMBoZ]

What is the 'tube sound'?

In my opinion it is this; A superior sense of coherency, vividness and saturation of sound that stems from electrons flowing through a vacuum as opposed to some other solid matter.

Is it fair to say, that typical tube amps are more interactive with speakers, than typical solid state amps? (Ignoring the reasons why)

Every amp you listen to 'music' from, can only be 'listened to' via speakers. The sound you here is a result of the amps interaction with woofers, interaction with any given TWEETER. A big part of the 'tube sound' is related to how valve amps interact with speakers, for better or for worse.

It is my opinion that there are certain musical instances that are better expressed or conveyed to the listener when these interactions occur. It is also fair to say that there are instances when the musical intent is compromised because of these interactions, but 'listening' to this will be entirely a subjective matter of taste.

Valve amps must do something or have some particularly unique quality that keeps them popular with discerning or enthusiastic music appreciators/listeners - despite all their other apparent flaws (distortions, thermal inefficiency etc. etc.) It isn't just that they glow, look funky etc.

 

[DF96]

In my opinion it is this; A superior sense of coherency, vividness and saturation of sound that stems from electrons flowing through a vacuum as opposed to some other solid matter.

I suspect that even if you limit yourself to the active devices you may find that the actual solid material distance travelled by electrons in a tube amp is greater than in an SS amp. I say this because cathode oxide coatings are probably thicker than BJT junctions, and tube internal wiring (e.g. from the anode) is longer than semiconductor bond wires.

In both types of device the controlling mechanism is electric fields. Exactly the same type of electrons are used - there is only one type to choose from.

I don't know what 'saturation' of sound means. Normally in electronics we try to avoid saturation because it often causes peak clipping.

Is it fair to say, that typical tube amps are more interactive with speakers, than typical solid state amps? (Ignoring the reasons why)

You would have to define 'interactive' but then you start getting dangerously close to looking at reasons and then before you know it you are back into the nasty world of engineering and physics and facts.

 

[Evenharmonics]

For you, maybe. But you have yet to define to what you are being Highly Faithful (Fidel). Without that, there is no goal, no reference.

Look here.

 

[TiMBoZ]

the actual solid material distance travelled by electrons in a tube amp is greater than in an SS amp

Ok, but isn't the 'electrons in vacumm' occurring in part that's critical to the sonics of the amp circuit? Like it's not just about distance travelled by electrons in a vacuum; it's the core part of the amp circuit, kind of thing?

 

[SY]

If charge doesn't travel through the solid parts, the sonics will be nonexistent.😀

 

[TiMBoZ]

Speaker interactions aside. What I'm trying to say, suggest or determine otherwise: is that the 'business of amplifying' occurs in the valve (across vacuum) or in the transistor (across solid matter - metal/silicone?). Is this is the source of the differences I hear. At present I believe so

 

[SY]

As DF96 pointed out, inside the valve (and there's nothing magic about that spot as opposed to all the other places that charge travels), the charge must traverse several different kinds of metal and metal oxides.

I think you mean "silicon" rather than "silicone." The latter has its own virtues...

 

[TiMBoZ]

But how significant to the 'sound of valve amp' - is that part of the circuit? How much of the 'business of amplifying' occurs at that part the circuit? My listening in the field suggests, that it's somewhat significant.

 

[DrewP]

As SY would know from a recent PM I have sent to him, I've just finisehd building an LED biased PP tube amplifier.

Relatively high output impedance, no global or local feedback to reduce distortions and triode connected the whole way through...

This amplifier does not sound overly sweet, soft, slow, wooly, flabby in the bass, euphonic or any of the other myriad attributes supposedly allocated to "tube sound" even though it does not have a low output impedance of a ruler flat frequency resonse or feedback to reduce its distortions...

Oh, and it exhibits hard clipping like a SS amp...

I guess "tube sound" must come from something other than what I've used...

 

[SY]

How much of the 'business of amplifying' occurs at that part the circuit?

As much as anywhere else. If electrons aren't emitted from the cathode, the tube doesn't amplify. Or do anything else, for that matter.

 

[fas42]

With regard to 'enhancing' the sound to improve the listening, this has never ever been my experience - faulty reproduction has always degraded the sound, for me: detail is lost, treble is harsh, defects of the recording itself overwhelm, get in the way of appreciating the musical 'message'. So my approach has always been to remove or resolve the issues which in any way "dirty" the reproduction process.

So how do I know this is the 'right' approach? Because, the end result is that every recording becomes satisfying to listen to, I'm able to subjectively hear past any intrinsic deficiencies of the recording itself and appreciate the music making while remaining relaxed, at ease, without stress. The analogy I've used a number of times is listening to a busker who happens to be a top notch musician, performing on a busy, noisy street: at least for me I have no problem switching off all the irrelevant, extraneous background racket and "zooming" in to the performance in front of me - which has the vitality, the verve, the magic of hearing the person perform in the very best, acoustically perfectly manicured auditorium.

That's how it works for me: the musical performance lives in a "different" acoustic space from that of the remaining distortion, and I have no trouble unconsciously separating those "places". A further benefit is that really "bad" recordings can be replayed at very high, natural sound levels with no discomfort, because the mind is able to do this 'filtering' with minimum conscious effort ...

 

[TiMBoZ]

What if electrons are emitted? Don't the electrons flow/jump from one plate to another through a vacuum?