The ups and downs of tube vs. SS...

What are the ups and downs of both?

In the end, its all about your investment. Tubes are hotter, and the power they make watt for watt is more expensive.

If the closest you can get to real music is your goal, tubes are the way to go. If its anything else (sound pressure, low heat), probably solid state.

The reason is simple: the ear perceives sound pressure via the presence of higher ordered harmonics- the 5th and above. It does **not** gauge the fundamental tones! This is probably due to the fact that pure fundamental tones are very rare in nature.

Tubes might seem to make more distortion when you look at amp specs on paper, but the harmonic distortion they make tends to be the lower orders- the 2nd - 4th, to which the ear is far less sensitive!

So a simple way to put it is that tubes more closely obey the most important rule of human hearing which is how we sense sound pressure. So it is much easier to build an amplifier that sounds natural to the human ear with tubes.

Another way to look at this is that the ear hears volume on a logarithmic scale.  In a way, it comes quite close to sensing harmonics on a log scale too- being so sensitive to the higher harmonics that they can easily be detected by the ear even when the THD is only 0.005%.

We don't weight the harmonic distortions electronics make under our current measurement regime. The 2nd order (which contributes to 'richness') is given the same importance as the 7th! This is quite odd, because we've known since the 1930s that the ear is more sensitive to higher orders (see the Radiotron Designer's Handbook, 3rd edition). In the 1960s, General Electric did a study confirming this, that showed that humans don't care if up to 30% of 2nd order is present, but object quite a lot of 0.1% of the 7th order is present. Yet despite having this knowledge, we continue to ignore the implications (probably because of the money; transistor amps are simply less expensive to make and often have higher profit margins).

This is why there is a solid state/tubes debate; its why tubes are still around after being declared 'obsolete' in the 1960s; and the debate won't be going away any time soon. When our testing systems continue to ignore some of the most important aspects of human hearing/perceptual rules, you can count on the spec sheet not telling you how the equipment sounds (or doesn't) just as spec sheets have done for the last 60 years.

I am a tube guy for quite while. I'm like to read articles concerning human hearing & perception of sound pressure levels and the effect of different levels of harmonic distortion on human hearing. Any suggestions?

There is precious little about this specific topic- for example GE's study is not online. But you don't need to read anything if you have the right test equipment- a sine/squarewave oscillator (as opposed to a generator), an amplifier of any kind, a speaker and a VU meter.

Set the oscillator to sine and drive the amp through the speaker. Put the VU meter in the circuit and set the level at 0VU. Then cover up the meter, and switch the oscillator to 'square' and set the level so it sounds at the same level as before. Uncover the meter, and you will see how much more sensitive our ears are to the higher ordered harmonics!

The tube amp sound less fatigue than SS and the impulse/dynamic is simply great. The only thing is that in the low end the energy is that strong as a SS.

If you want the bass energy in the tube amp to be right, the bass range of the speaker should be as high or higher than that of the mids and highs. Also, you will want to avoid 4 ohm speakers- with almost any output transformer, there is a loss of low frequency bandwidth driving from the 4 ohm tap as opposed to the 8 ohm tap (plus the amp will be smoother and more detailed driving higher impedances, which is true of **all** amplifiers). Also, keep your speaker cables short (under 7 feet!) and the connections at either end tight.

Two audio shows really shaped my opinions on tube gear. One was a single ended symposium put on by Stereophile mag in the early-mid 90's in Philadelphia and it was the first time I heard SET gear. That day I heard 845 amps, 300b, 211, and 2A3 amps from very well known designers. They were all incredible sounding but the one that caught my ear the most was the least expensive and lowest powered amps the FI 2A3 monoblocks. At 3 wpc, they were able to fill a 40 x 40 room with 104 db horn speakers.

@lou_setriodes , I'm betting that this show was actually the Philadelphia Triode show and it was more like about 1998 or so. I was at that show (and played our amps), and the speakers used were made by what is now known as Classic Audio Loudspeakers, and the Fi was indeed quite impressive. I thought it was the best sounding of all the SETs presented.

I drive some very inefficient electrostats that have impedance curves that drop to 1 ohm with big class A power amps.

You might be surprised to find that tube amps can manage a load like that. IME many people think ESLs are inefficient and difficult to drive but I find that isn't completely true- they are often more efficient than thought, and their load can be problematic for certain amps, depending on the ESL.

For example Sound Labs are hard to drive for solid state, as you need about 800 watts to keep up with a 200 watt tube amp on the same speaker. This is because the speaker is 30 ohms in the bass, which knocks the solid state amp down to about 200 watts, while a tube amp is unaffected. At the extreme top end a Sound Lab is 1.5 ohms and our amps (which don't do particularly well with low impedances as they are OTLs) don't have any problems with them.

1 ohm (if correct) is not crazy bad for a tube amp because it occurs at 20KHz where there is little energy. So you might want to try a tube amp on the speaker and see. OTOH, Martin Logan ESLs are typically about 0.5 ohms at 20KHz and no tube amp does them well (they sound rolled off in the highs) without a set of ZEROs (www.zeroimpedance.com) to boost the impedance.

My favorite is a Fisher 500C unrestored.

That won't last long if the filter capacitors don't get replaced. As they fail, they tend to take the power transformer with them. It can get pretty ugly as the transformer might leak nastiness when it gets really hot.

So get the caps replaced- and the amp will sound even better.

I was wondering about a tubed buffer stage between source and amp?
If one was to go with either SS, or tubes, will it make a huge difference? Or, is it just a waste of money?

If you don't need the gain a tube buffer is a good idea. You will loose some voltage with a tube buffer (as it will be a cathode follower); to minimize that and to decrease distortion a mu follower is not a bad idea. The reduced output impedance will allow you to run longer cables and will reduce the sonic qualities of the cable.

We've optionally built buffer circuits in our preamps that have direct-coupled balanced outputs.