Technically, what exactly is TUBEY sound?

[solderdude]

You've designed tube headphone amps. What led you to that?

Jeremy (garage 1217) asked me to help him with that after he read my mods for the Bravo V2 which objectively improved performance of these cheap (and crappy as well as flawed but well liked amps)
I don't use tubes myself but do like watching the glowy glass thingies despite their flaws.
With tubes it matters how you use them. (i.e. topology and parts around it) that can, or do not, 'colour' the sound in a way that is either appreciated by some or simply works admirably and cannot be distinguished in a blind test.

I also designed other headphone amps (SS) that measure better (Kameleon, C.H.A.M.P. TX-gain amp) and quite a few other headphone and speakers amps as well.

To answer the O.P. I will present my opinion:



FREQUENCY RESPONSE

May roll-off high frequency and could reduce bass extension (when transformers are used). Well made transformers roll-off outside the audible range. An amp has to have severe roll-off on both ends to come across as 'midrangy' and probably has some very cheap and poor transfomers in it. Tubes, like SS do not show a 'lift' in certain parts of the FR, transformers, poor designed and components around those parts can.


DISTORTION

Distortion is different in harmonic signature. With well designed tube amps distortion can be as low as SS amps (with enough tubes and feedback).
As the distortion becomes higher for larger voltage swings the bass frequencies have the most amounts of harmonics added while smaller amplitudes have lower levels. Because of this there are a bit more added harmonics for lower frequencies.
For that to become audible it would have to be substantial. Distortion numbers, for that reason say little. It is the characteristic of it.
Of course this does not apply to most decent and good tube amps.

Decay

For the decay to become different the linearity of the amp should be very poor for this to happen so large amplitudes would have to be compressed a LOT (would sound very poor). I think the decay thing is subjective mumbo jumbo and can not be shown to exist in reality.

CLIPPING

Hifi amps normally do not run into clipping. That said... very low power amps could well be driven into clipping when playing a little louder.
This is less 'harsh' and acts as a soft-limiter/compressor a bit and is less objectionable.
Add to that most tube amps supply power rather than act as a limited voltage source.
Because for most music most power is in the bass area and most speakers have an increase in impedance in the bass area a speaker amp that is rated say 30W can supply a higher voltage swing in the bass than a 30W SS can.
Power = (Voltage x Voltage) / impedance and as a tube amps provides power and the impedance is higher there can be a higher voltage on its terminals than for higher frequencies.
So a 30W tube amp can play louder than a 30W SS amp (depends on the speakers and topology)

SIGHTED

This one is left out but think this is the biggest reason for the perceived differences and preferences of most tube lovers.
Those running SET with high output resistances simply prefer a low DF( = Damping Factor see explanation in post #45) or might have a mental bias.
I am pretty convinced that this has a lot to do with the preference.

Long ago I had gotten a defective tube amp. As I was a bit scared for the high voltages didn't want to repair it nor did I know how tubes worked.
I did like the look... So I removed the bits inside, built the most basic (4 transistor) SS amp inside, used the metal frame and heatsink and connected the heaters. The 'anti-thump' output relay timing was extended to 1 minute.
Whenever I was in the opportunity to 'show off wonderful tube sound' I took it along and unleased it on unsuspecting listeners.
The few times I demonstrated it to 'hifi lovers' most of them heard the tube sound and loved it. One didn't like the sound and later mentioned he hated tube amps for their sound.

It was 'sighted' and not level matched. Just the imprint that it was a tube amp was enough.
Funnily enough a few years ago on a subjective site some people commented on a very similar amp design (one transistor for voltage amplification and PP output stage) how 'wonderfully tubey' that design sounded.
That would invalidate my 'secretly SS tube amp' findings to them IF they knew (never told anyone they actually heard a crappy SS design)

People don't like to be fooled (knowingly) but are generally asking/desiring to be.

Just anecdotal... no witnesses nor test results written down on paper.

 

[Hipper]

Solderdude, thanks for that interesting reply.

However I'm a bit of an ignoramus as far as amp designing is concerned. Could you please tell me, briefly, what 'Sighted' and 'DF' means?

 

[THW]

Solderdude, thanks for that interesting reply.

However I'm a bit of an ignoramus as far as amp designing is concerned. Could you please tell me, briefly, what 'Sighted' and 'DF' means?

a sighted comparison means you conduct said comparison with full knowledge on the subjects being compared.

this introduces psychological bias which compromises the accuracy and reliability of the comparison results.

(i hope i phrased that correctly and in a way people understand)

 

[Hipper]

a sighted comparison means you conduct said comparison with full knowledge on the subjects being compared.

this introduces psychological bias which compromises the accuracy and reliability of the comparison results.

(i hope i phrased that correctly and in a way people understand)

Thanks. At first that is what I thought it meant but then as I read on it seemed perhaps a more technical term!

 

[solderdude]

Sorry..

'Sighted' means one knows what is being used. That does not seem as a terrible thing. Almost everything one uses/evaluates is done 'sighted'.
One looks at it and made a purchase (or desires to) done on looks and info gathered earlier.
The problem with sighted is that there is a 'mindset' that, while one can deny it influences their decision making, it actually does.

Proof for this is the fact that when one does not know what is playing (there are ways for this) the differences one clearly heard suddenly don't appear to be there. And when they are (which is equally possible) they are large enough to be audible. In that case they also are very measurable.

This is true in the electrical field which is easier to measure than acoustical. In both cases we use music and our own ears.


DF is 'Damping Factor' a rather nonsensical term with numbers that do not work 'linear'.
DF is the factor between the load resistance (not impedance) of the load (in case of a speaker usually a substantially varying impedance) divided by the output resistance of an amplifier. Usually at 1kHz.

A DF > 10 is generally considered as sufficient. There is no practical difference between a DF of 100 and 100,000 but the numbers do differ immensly.

A high output resistance (low DF) can have an influence in the frequency response as well as resonances not being damped as well as the designer expects it to be 'damped' electrically. Most damping with speakers is (or should) be done acoustically (mechanically) and only a small portion of the total damping may have to come from electrical damping.

An unusal high output resistance thus affects the frequency response. The higher the output resistance and the bigger the impedance swing of a speaker the more 'severe' the change in frequency response is.
This can certainly reach audible levels with some speakers and high output Z amplifiers. OTL tube amps often are higher impedance. The same is true for current drive amplifiers (which aren't fully current drive but at some load turn into voltage drive amps for practical reasons).

 

[Krunok]

DF is the factor between the load resistance (not impedance) of the load (in case of a speaker usually a substantially varying impedance) divided by the output resistance of an amplifier. Usually at 1kHz.

I find it a little awkward to speak of resistance with non DC signals. Shouldn't we always talk about impedance with AC signal with frequency of 1 kHz, as neither amplifier output nor loudspeaker input will show pure resistance character with AC/1kHz conditions?

Unless of course DF is simply calculated using resistance only values? Is that the case?
For example, my RMB-1075 has DF 180 at 8 ohms. Does it mean that Zout value was 0,044 ohms so 8/0,044=180?

 

[solderdude]

Damping factor is measured with a resistance. The number actually varies when considering impedances.

The output resistance of an amp is as good as resistive at 1kHz. resistance can go up and become capacitive when output capacitor are used and go up when inductors are placed at the output as well as because of the feedback circuit itself.
Around 1kHz it behaves resistive. Most speakers will also be resistive in that area unless there is a crossover point.

For example, my RMB-1075 has DF 180 at 8 ohms. Does it mean that Zout value was 0,044 ohms so 8/0,044=180?

Yes that is exactly what it means. But it could differ at the lowest frequencies or at very high frequencies. The DF will be smaller there.
It will also be lower at lower impedances so 90 at 4 Ohms

It would be preferable if manufacturers would just mention the output resistance but this could confuse customers as output R and max output current are not linked at all but people could draw that conclusion.
The DF number is more practical .. > 10 is O.K. > 40 is already perfect (around 1kHz)

 

[LaLaLard]

Many say harmonic distortion has euphonic qualities like added depth or density. My experience has been that all it adds is indistinctness.

Which masks some inferior recording(good in some aspect but no where near definition of HiFi anymore)

 

[LaLaLard]

Jeremy (garage 1217) asked me to help him with that after he read my mods for the Bravo V2 which objectively improved performance of these cheap (and crappy as well as flawed but well liked amps)
I don't use tubes myself but do like watching the glowy glass thingies despite their flaws.
With tubes it matters how you use them. (i.e. topology and parts around it) that can, or do not, 'colour' the sound in a way that is either appreciated by some or simply works admirably and cannot be distinguished in a blind test.

I also designed other headphone amps (SS) that measure better (Kameleon, C.H.A.M.P. TX-gain amp) and quite a few other headphone and speakers amps as well.

To answer the O.P. I will present my opinion:



FREQUENCY RESPONSE

May roll-off high frequency and could reduce bass extension (when transformers are used). Well made transformers roll-off outside the audible range. An amp has to have severe roll-off on both ends to come across as 'midrangy' and probably has some very cheap and poor transfomers in it. Tubes, like SS do not show a 'lift' in certain parts of the FR, transformers, poor designed and components around those parts can.


DISTORTION

Distortion is different in harmonic signature. With well designed tube amps distortion can be as low as SS amps (with enough tubes and feedback).
As the distortion becomes higher for larger voltage swings the bass frequencies have the most amounts of harmonics added while smaller amplitudes have lower levels. Because of this there are a bit more added harmonics for lower frequencies.
For that to become audible it would have to be substantial. Distortion numbers, for that reason say little. It is the characteristic of it.
Of course this does not apply to most decent and good tube amps.

Decay

For the decay to become different the linearity of the amp should be very poor for this to happen so large amplitudes would have to be compressed a LOT (would sound very poor). I think the decay thing is subjective mumbo jumbo and can not be shown to exist in reality.

CLIPPING

Hifi amps normally do not run into clipping. That said... very low power amps could well be driven into clipping when playing a little louder.
This is less 'harsh' and acts as a soft-limiter/compressor a bit and is less objectionable.
Add to that most tube amps supply power rather than act as a limited voltage source.
Because for most music most power is in the bass area and most speakers have an increase in impedance in the bass area a speaker amp that is rated say 30W can supply a higher voltage swing in the bass than a 30W SS can.
Power = (Voltage x Voltage) / impedance and as a tube amps provides power and the impedance is higher there can be a higher voltage on its terminals than for higher frequencies.
So a 30W tube amp can play louder than a 30W SS amp (depends on the speakers and topology)

SIGHTED

This one is left out but think this is the biggest reason for the perceived differences and preferences of most tube lovers.
Those running SET with high output resistances simply prefer a low DF( = Damping Factor see explanation in post #45) or might have a mental bias.
I am pretty convinced that this has a lot to do with the preference.

Long ago I had gotten a defective tube amp. As I was a bit scared for the high voltages didn't want to repair it nor did I know how tubes worked.
I did like the look... So I removed the bits inside, built the most basic (4 transistor) SS amp inside, used the metal frame and heatsink and connected the heaters. The 'anti-thump' output relay timing was extended to 1 minute.
Whenever I was in the opportunity to 'show off wonderful tube sound' I took it along and unleased it on unsuspecting listeners.
The few times I demonstrated it to 'hifi lovers' most of them heard the tube sound and loved it. One didn't like the sound and later mentioned he hated tube amps for their sound.

It was 'sighted' and not level matched. Just the imprint that it was a tube amp was enough.
Funnily enough a few years ago on a subjective site some people commented on a very similar amp design (one transistor for voltage amplification and PP output stage) how 'wonderfully tubey' that design sounded.
That would invalidate my 'secretly SS tube amp' findings to them IF they knew (never told anyone they actually heard a crappy SS design)

People don't like to be fooled (knowingly) but are generally asking/desiring to be.

Just anecdotal... no witnesses nor test results written down on paper.

You made my day Dr.Peng. Funnily tho I was thinking about exactly the same test a few months ago however I do not have the skill set to do such DIY otherwise I would love to see people’s face when i yank off the light bulbs on amp right after their compliments on the wonderful tubey sound and the music goes on...without all the “tubes”...

 

[Sal1950]

Tarantino

If your jazzed to see Once Upon A Time, I'd think twice.
It might not be at all what you think, Tarantino is at it again re-writing history to ridiculous extents like he did for Inglorious Bastards.
Just don't say I didn't warn ya.

Seems I'm a lucky guy - just 10 km.

Yep mines just a 15 minute ride down the road, and it has a full bar too.

True, and Avatar, TRON is on the very short list of good 3D movies. To bad that I don't own a 4k to do 3D on. Have been watching music movies lately.
Rock of ages in DTS HD Master, Celine dion in Dolby True HD , Rhianna DTS HD Master (not that good mix) and Nightwish in 5.1 PCM

Best 3D films I saw were in the 1950s. I remember House of Wax with Vince Price, the barker outside the theater with the rubber ball on a paddle had people in the audience ducking as the ball came right off the screen.

 

[watchnerd]

Which masks some inferior recording(good in some aspect but no where near definition of HiFi anymore)

I would posit that it has been at least a decade, maybe two or more, since high fidelity was the driving motivator of high end audio sales.

Vinyl isn't.

Tubes aren't.

High resolution digital audio is overkill for fidelity, but sells anyway.

MQA....well, let's not go there, shall we?

 

[Wombat]

If your jazzed to see Once Upon A Time, I'd think twice.
It might not be at all what you think, Tarantino is at it again re-writing history to ridiculous extents like he did for Inglorious Bastards.
Just don't say I didn't warn ya.


Yep mines just a 15 minute ride down the road, and it has a full bar too.


Best 3D films I saw were in the 1950s. I remember House of Wax with Vince Price, the barker outside the theater with the rubber ball on a paddle had people in the audience ducking as the ball came right off the screen.

View attachment 30363

I thought Inglorious Bastards was a satire on typical Hollywood war movies.

 

[watchnerd]

I thought Inglorious Bastards was a satire on typical Hollywood war movies.

Yeah, me, too....

Who expects any Tarantino movie to be historically accurate?

They're all homage / satire pieces.

 

[RayDunzl]

If your jazzed to see Once Upon A Time, I'd think twice.

Found that Tuesdays are the least expensive day to go.

Not by much, get the old fart discount on the regular days.

Maybe I'll do some reconnaissance to see when it is least busy vs time I want to go.

Maybe next week.

 

[Wombat]

A slight variation on the topic:

Old time tubey sound:

Edison


Ships: https://en.wikipedia.org/wiki/Speaking_tube

Example

From an esteemed audio man.

 

[Burning Sounds]

If you use JRIver or another player that can use VST plugs-ins this is a free tube emulator - Wave Arts Vintage tube saturator

Running it through REW it seems the drive knob increases even order harmonics.

They claim it is the most accurate tube saturator available and is modelled on an actual circuit (2 12AX7s) and uses a Baxandall 3 band EQ.

How much of their claim is true I don't know (or really care), but I did find it fun to play with the various presets. The basic tube warmth setting (no EQ) is subtle, but audible. A +1 or 2 db setting on the midband eq moves vocals forward, and is not unpleasant if not overdone. No longer accurate to the original recording of course, but hey it's free and one click switches it out.

There's a review of it on Tape Op here.

 

[Pluto]

I would have preferred other people there

Indeed. Not a film I would like to watch in a large, open space when you don't know what might be creeping up, out of the darkness, behind you...

 

[Pluto]

I would like to know what is the scientific reason behind the "attractiveness" of tubey sound

See this post

 

[jsrtheta]

Tube sound means "bright, forward, alive" and "laid back, mellow, and soft."

As well as "veiled, yet transparent".

 

[LaLaLard]

See this post

Read and liked