If "Tube Sound" Is a Myth, Why Tubes?

[mhardy6647]

That article starts off with nonsense.

It doesn't just start off with nonsense

 

[Rzezniq]

It doesn't just start off with nonsense

Well you would think that someone with 50 years of experience in audio knows what they are writing about. But since I'm not a subject expert I would appreciate if you could point all the nonsense so I can email the author.

 

[dfuller]

It doesn't just start off with nonsense

I more meant that the entire premise of the article is nonsense.

 

[Rzezniq]

I more meant that the entire premise of the article is nonsense.

Care to elaborate?

 

[mhardy6647]

sure -- give me/us some time. There's some pretty vague and inaccurate statements in the stuff quoted. Now, perhaps the author was just trying to "write down" to his/her/their audience, but if so, he/she/they didn't do a very responsible job. We see that in science reportage in the mass media quite a bit (e.g.) -- but that's a different story.

Bob Carver's been fiddlin' and diddlin' with ss and vacuum tube audio for a long, long time, too -- but it doesn't stop him from spouting drivel (or, at least, allowing his name to be associated with drivel). Similar comments could be directed, I'd posit, towards ASR's poster child for pseudo-science in hifi, Paul McGowan. His videos are chockablock with statements that are just plain ol' wrong (or, at the least, antithetical towards with respect to prevailing wisdom in sound reproduction/audio engineering).

 

[mhardy6647]

Here's some low hanging fruit.

Regardless of how much Negative feedback is applied it is not possible for a valve amp to achieve a Zero output Impedance (100% damping factor) similar to a solid-state amp.

So are there in fact practical (e.g., commercially available) amplifiers (ss) with identically zero output impedance? This would represent infinite damping factot Damping factor is the ratio of load to source impedance: 100% damping factor is, at best, a misnomer. At worst, it suggests the author has no idea what damping factor means.

 

[Frank Dernie]

I more meant that the entire premise of the article is nonsense.

Well you would think that someone with 50 years of experience in audio knows what they are writing about. But since I'm not a subject expert I would appreciate if you could point all the nonsense so I can email the author.

Long experience does not guarantee understanding IME.
I have worked with people who have observed what happens in our field (motor racing) and attributed completely technically non-sequitur explanations to why. It tends to depend on early schooling and depth of physics understanding.
Not only that but some, one of the best designers I know, who was trained as a mechanic, had completely bizarre ideas about how some stuff actually worked, so some ideas he came up with were brilliant and some did not work at all.
Some of the beliefs are real moments but confidently held and defended.

 

[DonH56]

Too many errors and inaccurate assumptions to bother to try to correct... The author is clearly biased toward tubes and misrepresents (to put it nicely) many of the differences in devices, designs, and operation of tube (valve) and solid-state amplifiers. He also appears to share the myth that all feedback is bad, and I am not sure how feedback "quieten"s an amplifier. Negative feedback is an apt engineering or controls physics description of how it is applied but too many seem to think "negative" implies something bad ("negative") rather than simply being opposite in polarity.

 

[mhardy6647]

Yeah, quieten sounds like some weird transliteration from middle English -- or German.
I guess I should be thankful that the author didn't go on about dampening factor -- 'cause I hate it when my amplifiers get wet.

 

[mhardy6647]

So, @Rzezniq, if you're looking for some good, solid information on vacuum tube audio -- or an audio primer in general -- you could do far worse than spend some quality time reading Norman Crowhurst's trilogy on the topic, aimed at the technically curious, solidly-educated, but non-specialist "hifi hobbyist" of the late 1950s.
These are available as PDFs from a couple of different sources; I'll reference Pete Millett's collection, because it is an outstanding resource for folks interested in this kind of stuff.

http://tubebooks.org/Books/crowhurst_basic_1.pdf
http://tubebooks.org/Books/crowhurst_basic_2.pdf
http://tubebooks.org/Books/crowhurst_basic_3.pdf

source: http://tubebooks.org/technical_books_online.htm

The other astounding resource of technical information (and much, much, much more) on audio/radio/broadcasting, of course, is:
https://worldradiohistory.com/

Here are a few specific archives worth browsing and, ultimately, perusing:
https://worldradiohistory.com/Archive-All-Audio/Audio-Magazine.htm
https://worldradiohistory.com/BOOKSHELF-ARH/Bookshelf_Sams.htm
https://worldradiohistory.com/BOOKSHELF-ARH/Bookshelf_RIder.htm

If you're generally naive of basic electronics, you might also want to spend time with (ideally an earlier version) of the US Navy's "NEETS" program. There are many online resources for NEETS, here's one to get started with. NEETS starts simply and covers the bases thoroughly.
http://electriciantraining.tpub.com/

Fortunately for non-EEs like me, audio electronics (i.e., audio frequency amplification) is relatively basic and straightforward in the context of the whole pantheon of electronics (FWIW).

 

[Helicopter]

Long experience does not guarantee understanding IME.
I have worked with people who have observed what happens in our field (motor racing) and attributed completely technically non-sequitur explanations to why. It tends to depend on early schooling and depth of physics understanding.
Not only that but some, one of the best designers I know, who was trained as a mechanic, had completely bizarre ideas about how some stuff actually worked, so some ideas he came up with were brilliant and some did not work at all.
Some of the beliefs are real moments but confidently held and defended.

When I went through training to become a helicopter mechanic, some of the trainers had similarly bizarre ideas. After they invited me to the front of the class a couple times for algebra and physics lessons, everyone in the class pretty much agreed their explanations were better suited to the audience because they were accessible, though admittedly fantastic. I earned the nickname 'molecular structure' from the boss.

 

[Rzezniq]

Thank you for all the links. I'm always keen to learn. I think that this article is not aimed towards engineers and is trying to explain the difference between both amps in a way that a layman like myself can understand.

My question is:

Is the core assumption of Valve amps supplying the full power regardless of speaker changing impedance incorrect? As that's the core difference that I got?

 

[Inner Space]

Some of the beliefs are real moments but confidently held and defended.

Interesting. I found the same kind of thing in studios. E.g. I knew a guy who demanded a specific brand of razor blade to splice 1/4" tape. He would send runners out on urgent missions to the drugstore. He said his brand made for better sounding tapes. But ultimately I concluded he was semi-kidding ... searching for a token amount of personal control and agency, in a technical landscape utterly controlled by facts and reality.

Equally - slightly off-topic but germane to our type of discussions - there were mixers and masterers who literally didn't know and didn't care what equipment they were using. I knew a multiple Grammy-winning engineer who couldn't have told you what monitors he worked on. He was staring at them 10 or 12 hours a day and the brand simply didn't register. That's how it was for non-superstar engineers, freelancers especially - it was like being a truck driver, showing up at the depot, getting handed a set of keys. You got what you got. The pickiness audiophiles show would have been regarded as weird.

 

[Julf]

Is the core assumption of Valve amps supplying the full power regardless of speaker changing impedance incorrect?

Yes.

 

[mhardy6647]

Thank you for all the links. I'm always keen to learn. I think that this article is not aimed towards engineers and is trying to explain the difference between both amps in a way that a layman like myself can understand.

My question is:

Is the core assumption of Valve amps supplying the full power regardless of speaker changing impedance incorrect? As that's the core difference that I got?

(EDIT -- oops I read correct but you 'said' incorrect)
Yes, it is incorrect -- although the use of a transformer to couple the high impedance of the power output tube's plate to the low impedance of a loudspeaker load helps a lot in that regard -- which is why McIntosh uses autoformers (which function like a transformer, albeit without DC isolation) to this day in some of their solid state amplifiers.

Transformers and autoformers are amazing things, but they are imperfect (if nearly magical!) electromagnetic devices. As a consequence, designing and constructing one that has broad, flat bandwidth and can handle reasonable amounts of power across that bandwidth is 1) nontrivial and 2) not inexpensive. They have largely fallen out of favor for impedance coupling (especially in the audio power amplifier output stage) due to the inevitable compromises they present and due to the cost and weight that they add to a power amplifier.

In fact, the high output impedance (low damping factor) of a vacuum tube amp is why they are very sensitive indeed to the impedance curve of the loudspeakers one uses with any given vacuum tube amplifier. Much of the vaunted "system synergy" (and, shall we say, reverse synergy) arises from this, I think. The output impedance and "flat" frequency bandwidth of a vacuum tube amplifier can indeed be improved by judicious use of negative feedback -- so, there's really no good reason to fear or berate the application of NFB in any amplifier, as long as it's employed correctly!

 

[Helicopter]

(EDIT -- oops I read correct but you 'said' incorrect)
Yes, it is incorrect -- although the use of a transformer to couple the high impedance of the power output tube's plate to the low impedance of a loudspeaker load helps a lot in that regard -- which is why McIntosh uses autoformers (which function like a transformer, albeit without DC isolation) to this day in some of their solid state amplifiers.

Transformers and autoformers are amazing things, but they are imperfect (if nearly magical!) electromagnetic devices. As a consequence, designing and constructing one that has broad, flat bandwidth and can handle reasonable amounts of power across that bandwidth is 1) nontrivial and 2) not inexpensive. They have largely fallen out of favor for impedance coupling (especially in the audio power amplifier output stage) due to the inevitable compromises they present and due to the cost and weight that they add to a power amplifier.

In fact, the high output impedance (low damping factor) of a vacuum tube amp is why they are very sensitive indeed to the impedance curve of the loudspeakers one uses with any given vacuum tube amplifier. Much of the vaunted "system synergy" (and, shall we say, reverse synergy) arises from this, I think. The output impedance and "flat" frequency bandwidth of a vacuum tube amplifier can indeed be improved by judicious use of negative feedback -- so, there's really no good reason to fear or berate the application of NFB in any amplifier, as long as it's employed correctly!

Agree. My main tube amp synergizes with my vintage Klipsch speakers and not with my Focal Aria 948s. In other words, the amp breaks down and cannot perform properly with the impedance of the Focals, but functions properly with the Klipsch speakers.

 

[Rzezniq]

(EDIT -- oops I read correct but you 'said' incorrect)
Yes, it is incorrect -- although the use of a transformer to couple the high impedance of the power output tube's plate to the low impedance of a loudspeaker load helps a lot in that regard -- which is why McIntosh uses autoformers (which function like a transformer, albeit without DC isolation) to this day in some of their solid state amplifiers.

Transformers and autoformers are amazing things, but they are imperfect (if nearly magical!) electromagnetic devices. As a consequence, designing and constructing one that has broad, flat bandwidth and can handle reasonable amounts of power across that bandwidth is 1) nontrivial and 2) not inexpensive. They have largely fallen out of favor for impedance coupling (especially in the audio power amplifier output stage) due to the inevitable compromises they present and due to the cost and weight that they add to a power amplifier.

In fact, the high output impedance (low damping factor) of a vacuum tube amp is why they are very sensitive indeed to the impedance curve of the loudspeakers one uses with any given vacuum tube amplifier. Much of the vaunted "system synergy" (and, shall we say, reverse synergy) arises from this, I think. The output impedance and "flat" frequency bandwidth of a vacuum tube amplifier can indeed be improved by judicious use of negative feedback -- so, there's really no good reason to fear or berate the application of NFB in any amplifier, as long as it's employed correctly!

Sorry I meant correct and the whole thing went very meta.

Ok so to make things clear: will a valve amp in ultra linear mode or tetrode configuration (I think it's what author mentioned) supply more power to the speaker in line with its raising impedance? Is it any different to a solid state without autoformers like for example my Sony av receiver? I'm here to learn. Thanks for your replies so far.

 

[MakeMineVinyl]

Sorry I meant correct and the whole thing went very meta.

Ok so to make things clear: will a valve amp in ultra linear mode or tetrode configuration (I think it's what author mentioned) supply more power to the speaker in line with its raising impedance? Is it any different to a solid state without autoformers like for example my Sony av receiver? I'm here to learn. Thanks for your replies so far.

The short answer is no. A solid state amplifier delivers more power with decreasing impedance up to a point.

 

[mhardy6647]

Since we've drifted onto the topic of transformer coupling -- it is (of course!) perfectly possible to construct an output transformerless (OTL) audio power amplifier by employing a series-parallel array of power output tubes (ideally using a type of tube with relatively low plate impedance). Atma-Sphere (Ralph Karsten), e.g., has sold OTL amplifiers for quite some time, and Bruce Rozenblit (Transcendent Electronics) offers a smattering of OTLs as kits.

Whether this is a good idea or not is... a matter of debate. If nothing else, having nothing to protect a relatively delicate & expensive loudspeaker from the high DC plate voltage in an OTL if anything goes wrong is enough to keep my interest in OTLs as academic only.

http://www.atma-sphere.com/en/index.html

Here's a nice write-up of the dual power triode tube used in Atma-Sphere's amps. The 6AS7 is (sort of, at any rate) similar to two 2A3 power triodes in one envelope. http://www.atma-sphere.com/en/resources-6as7g.html

Rozenblit aims a little lower (in power and price!) and has been a little more experimental in his OTL configurations (at least in terms of some of the tubes he's employed) over the years, I'd opine.
https://www.transcendentsound.com/index.html

 

[mhardy6647]

Sorry I meant correct and the whole thing went very meta.

Ok so to make things clear: will a valve amp in ultra linear mode or tetrode configuration (I think it's what author mentioned) supply more power to the speaker in line with its raising impedance? Is it any different to a solid state without autoformers like for example my Sony av receiver? I'm here to learn. Thanks for your replies so far.

No. Simply put, the power delivered into a purely resistive load from a transformer-coupled audio power amp will not vary from tap to tap.
If the amp delivers 50 watts into a 4 ohm load through the OPT's 4 ohm tap, it will also deliver 50 watts into an 8 ohm load (via the 8 ohm tap) and... 50 watts into a 16 ohm load (via, you guessed it, the 16 ohm tap). The amp will work a bit more efficiently into the higher impedance load, because it makes better use of the secondary (winding) of the OPT, though.

In the real world, with a real loudspeaker load (i.e., an impedance curve that varies with audio frequency), things won't be so simple.

Here's the impedance (and phase angle) curve of an actual loudspeaker, the modern Polk Audio L-100 (a small two-way speaker).

source: https://www.stereophile.com/content/polk-legend-l100-loudspeaker-measurements

Polk calls the nominal impedance of the L-100 "3 to 4 ohms". In fact, it varies from a low of about 3 ohms @ ca. 5 or 6 kHz to a high somwhere (well) north of 20 ohms.
The "curve" of a truly noninductive (i.e., purely resistive) 8-ohm power resistor on the same chart would be a straight horizontal line at 8.0000 (Ohms).

See the "challenge"?