How can dirty amps sound so incredible?

A broad question I realize., but it's interesting as heck in my opinion. It's probably been the most interesting question in audio for me for 5 years now.

As an example the 300B pelted VAC 70/70 Ren is a magical amp. Depending on how its output impedance is matched with speakers, it can sound magical, haunting, sweet, soft, warm. Always airy. I used to call mine "enchanting", because it gave me and my friends goosebumps. Strange and hard to believe an amp could do what it did.

I've heard a similar sound from other 300B amps like Cary's 300B offerings. There are a great many other amps using these tubes or others with similar characteristics (the list of characteristics is long).

The VAC Rens aren't "faithful" electronically, regardless of how you define that word. And yet many of us (myself included) believe the sound produced by this amp can and often is as realistic as we've ever heard.

I assume that gods like Kevin Hayes will never reveal their innermost secrets, so I'm left to ask others. How do these amps sound so incredible?

 

I'm thinking it's the output transformers. I totally get what you mean though- I love the Manley Mahi monoblocks- EL84 based- they sound great. They also measure poorly.

 

A relatively simple circuit (usually), class A triode with no global feedback. Poor damping factor, and high even harmonic distortion compared to a more mainstream amplifier.

jeff

 

@vinylkid58 in the case of the VAC Rens, I'm really curious what input impedance the 300Bs "see" across the range for specific speaker input impedance. I still can't get over what I heard with my Ren 70/70 driving Joseph Audio Pulsars.

I've been too lazy to do the math here, admittedly. I was driving the Pulsars with the 8 ohm tap and feedback knob at zero ( I assume that means zero feedback, but I kind of doubt it). Whatever the numbers were, it sounded incredible.

BTW - the poor damping was evident and was the reason for selling the Ren even though I have to wonder if with the right speakers ...

 

Distortion is pleasing if it's the right kind.

 

Some speakers were designed to partner with amps like this, just like many modern speakers are designed to work with SS amps with lots of negative feedback. Do your research and don't get rid of the baby with the bathwater.

 

It's the "magic" of a good tube amp and properly matched speakers that give an amazing impression of realism, despite the relatively poor electronic output measurements. Thus some solid state fans will scoff and say it's the tubes "colorizing" the sound that makes it sound that way -- "realistic" that is. I've always found that ironic, that although it sounds most "realistic", it's actually not as "accurate"! Huh?

 

A thirty year old Honda is more reliable and powerful than a new Harley-Davidson. A $20 quartz is 100x more accurate than a $30,000 Swiss mechanical. Evan Williams gets you just as drunk as Louis XIII.

 

That's certainly true, and makes me realize I wasn't being quite honest - I had to throw out the baby before I sunk $1000-$1500 on new 300Bs. But it would have gone anyhow even though I held onto it long enough to find that it did sound great for some speakers. (like magnepans). But since I had about 4 other amps sitting around (MX-Rs, Einstein, Music Reference) something had to give.

Maybe most important is this: I can get another one or even a 30/30 if I want later. I know others who have done the same as me with the Rens and admit it was wonderful, but still got rid of theirs knowing they could always get another one later.

 

Actually, the 70/70 sounded great with every speaker I owned because I didn't own any that were terribly difficult to drive. But to @layman's point, if you choose to stay with this amp there are a good many speakers that would sound excellent regardless of the effect the damping factor had on less appropriate speakers. Point well taken.

 

It's all about voicing. Circuits with a lot of character (distortion) require a lot of artistic judgement to get the sound right; that's what someone like Kevin Hayes is good at. Places like JBL/Harman frown on this method of circuit engineering because it gets hard to get repeatable results week after week, year after year.... when manufacturing.

 

Right. I'm sure Kevin had a lot of fun dialing things in while voicing. I just don't understand how just the right amount of distortion here and there, along with lack of flat response in a certain range can lead to images hovering 4 feet further back than they do with other amps. Spooky as heck with my Pulsars And Vandy Quatros. I think that was the biggest surprise for me: that depth and air was always there more or less for different speakers. But always there.

That first day I heard it I was certain there was some sort of electronic delay mechanism together with huge amounts of cross-talk. Obviously I was completely wrong and have remained dumbfounded.

 

Vintage dirt!

 

A $20 quartz is 100 times more accurate than a $30k Swiss mechanical? Let’s see proof.

 

There are a lot of tube amps which break every single one of those rules (except for poor damping factor), and yet which still sound superb. So I don't really think that this is a required recipe for great tube sound.

 

Quartz accuracy report
https://tf.nist.gov/general/pdf/2276.pdf

COSC requirements
COSC - Wikipedia

So the average quartz is only 33.3333333x more accurate. My bad!

But is this level of nitpicking really the lesson you’re taking from my earlier post? The point is simply you cannot base everything on spec. That’s just bench-racing

 

Sure there is, you have to actually have the tube in the circuit.

 

I was just curious if you could prove that statement or not. I have two Swiss mechanical chronometers that will keep to +6 seconds per month as long as I wear them and thereby keep them wound up. One is an Omega Seamaster Pro that only cost me $1700 new 14 yrs ago and the other is a Rolex I bought new for $5000 10 yrs ago. That’s a lot less than $30k.

 

Simple, designed by people who used simpler circuity, for people who listen to music instead of spec sheets. This principle even applies to solid state amplifiers. The older, pre IC based, simpler circuitry designed closer to tube era principles in good order, tends to be sweeter sounding (efficient speakers need to be used), than the overdesigned, overdamped, spec sheet friendly designs which followed. Listen first. I feel like you listened for yourself, then decided with your ears. Therefore, you found the sound you liked. You can overdesign an amplifier, tube or transistorized. And suck the musicality out of it. McLover's law, get the midrange and the musicality right, the rest follows.

 

I’m glad your watches are working above average for you, that’s great.

The point, however, is that people will spend ten times as much or more on a Hublot (or a PP, AP, VC, ALS, JLC, GF, etc) and not have a more accurate watch than your Seamaster, or a Citizen, or a Timex. People can like goods that are not always the best performing if they like them. Like OP’s less-than-linear amp.

 

Every guitar slinger knows a little, or a lot of, distortion is a good thing.

 

Just about every smart phone is more accurate since they are synchronized against a network of computer clocks around the world, which in turn are synchronized against the world's network of atomic clocks. Atomic clock accuracy is mind blowing. They have designed an atomic clock that is accurate to +/- one second over the lifetime of the universe.

To put it into context. The time on my smart phone is correct. It will still be correct in a years time. Should it still be working in ten years time, it will still show the correct time. Had the earth not been destroyed by its sun, my smart phone would probably still show the correct time in 15 billions years time. Cheating I know, but that's modern techology.

 

Thanks I needed that laugh. In five years that smartphone will in all likelihood be an obsolete paper weight. My crummy Swiss mechanical will be around decades after I’m gone, ticking away the time in good order and it’s value will keep going up.

 

Right, but whatever smart phone you have, it will keep time better than your Swiss watch (or any other watch besides, say, an Apple watch) because it is synchronized against the best clocks in the world.

Note: I did just have to force my computer to synchronize because I noticed it was running a few seconds fast. It is now in synch with the USNO clock.

 

The first comment that I will make is, who in the first place decided that "measurements" had anything to do with realistic sound?

NO ONE!

They don't and never did, the whole measurement's thing is BS created by the manufacturer's.

None of this has anything at all to do with how something is going to sound anymore than a chemical analysis of raw ingredients will provide even a clue how something will taste.

I might consider some measurements, such as size, power output, circuit type... in order to get some idea what the animal is. But just knowing what a breed of dog is on paper will not tell you the personality of the dog.

You can have your gene's mapped, what does that tell you other than you have blue eyes and blond hair...?

OF all of the equipment that I have DAC's, preamps, phono preamps, CD's, processor's, turntables, cartridges amplifier's and speakers, I have no idea how any of them "measure". I really could care less. I care how they sound and I don't so much really care about that.

What is important to me is synergy, total system synergy. I don't care how something sounds by itself, I care how it sounds in my system as a whole.

Does it belong there or not? Is is a good fit?

Another thing that is thrown around is tube "distortion". Time to clarify that a bit.

Take a piano, and play a middle "A" at 440-cycles. that is known and the fundamental frequency, or first harmonic. In nature, there will be a doubling of that frequency, which will be the second harmonic, which is 880-cycles, which also happens to be the frequency of the "A" in the next octave.

Here are the octave's on a piano.



Here are the corresponding frequency's.



The "A" in the next octave up is at 1760-cycles. and so on. We hear in octaves. our musical instruments have the fundamental frequency and then a doubling of that frequency and then doubling of that frequency and so on. This is known as harmonics. But harmonics need a better explanation.

Harmonics are multiples of a fundamental frequency (or 1st harmonic).

So the harmonics of 440-cycles are 440, 880, 1,320, 1760, 2,200, 2,640, 3,080, 3,520...

Let's break this down a bit...

440 - 1st harmonic "A"
880 - 2nd harmonic "A"
1,320 - 3rd harmonic
1760 - 4th harmonic "A"
2,200 - 5th harmonic
2,640 - 6th harmonic
3,080 - 7th harmonic
3,520 - 8th harmonic 'A"

When a piano string vibrates, it creates the 2nd harmonic, the 4th harmonic, the 8th harmonic.

These are natural with instruments and are pleasant for us to hear.

All amplifier's produce harmonic distortion, more so, if over driven.

SS amplifier's by their nature produce more distortion in the odd number of harmonics. While tubes, by their nature, produce more distortion in the even number of harmonics.

Tube harmonic distortion is more natural to our human ears Which is why we can tolerate more tube distortion, and not only does it not bother us, it can make the music sound more full and natural.

Odd harmonics sound harsh and unnatural to us.

If a SS amplifier is driven into clipping by a transient, the result is no longer a sign wave but a harsh sounding square wave.

When a tube amplifier is over driven, the sound becomes "mushy".

While neither are necessarily desirable, mushy is not as offensive to our ears as harsh.

Most of the SS "heads" that throw around terms like tube distortion have no understanding of what they are talking about. There are the individuals who elevate SS and put down tubes. Judge them for what their understanding is (or isn't).

Just some thoughts on measurements, dirty tube amps and distortion...