That "tube sound" and power ratings

As I understand the question, you are seeking to determine if in general it makes sense to prefer a lower powered tube amp to a higher powered tube amp (assuming that both amps can provide sufficient watts for the application), due to an expectation that the lower powered amp may, under typical conditions, tend to distort more than the higher powered amp. And thereby provide sonic characteristics that are more "tubey" and less solid state-like than would otherwise be the case.

That is a logical question, but I think it focuses MUCH too narrowly on one single factor in tube amp sonics for any answers to be useful. There are simply way too many variables involved in the design and the performance characteristics of an amplifier, and in its interaction with the technical characteristics of whatever speaker is being used, for a useful answer to be defined. Certainly among different makes of amplifiers. And even when considering amps from a given manufacturer having different power ratings, that are otherwise generally similar in design, I see no reason to expect much consistency among the product lines of different manufacturers as to whether the higher powered or lower powered amp would be preferable. Or even much consistency across a variety of speakers when a comparison is performed between a given higher powered and lower powered amp, as a given tube amp can sound quite different depending on the magnitudes and phase angles of the impedances of the speakers it may be used with, and especially on how those parameters vary as a function of frequency.

Also, the question focuses on just one kind of distortion, harmonic distortion. There are other kinds, e.g., intermodulation distortion, transient intermodulation distortion, crossover distortion, etc. Some of these are neither generally specified nor even measurable in a standardized manner. And all involve a complex signal-dependent mix of different effects and different degrees of effects at different frequencies.

What I would suggest as the most useful line of inquiry, when you reach that point, is that you start a thread indicating the particular speakers that would be used, your budget range for amplification, the room size and listening distance, some idea of the volume levels you prefer, and (especially) whether or not your listening includes recordings having particularly wide dynamic range (i.e., particularly large DIFFERENCES between the volumes of the loudest and the softest notes, such as is often the case with well engineered, minimally compressed, classical symphonic music). The latter factor will dramatically affect how much power you need. Providing that information will get you many knowledgeable and useful suggestions from the members here, from which you can develop a “short list” of candidates.

Good luck. Regards,
-- Al

Great points by Ralph, as usual, with which I agree.

09-25-14: Charles1dad
Ralph, there's something that doesn't make sense concerning your comments about speaker designers. These designers have to have at the very least a rudimentary understanding of science and engineering. Something as basic as Kirchoff's law has to be easily comprehended and applied. They have the equipment available to confirm that concept and its effect. I just don't understand why you would be met with such resistance to something so plain and fundamental. How do these designers counter your point/position?

Charles, pending Ralph's answer, as I see it the key to reconciling that is the fact that the majority of speakers are designed with solid state amplification in mind. And with a few unusual exceptions, nearly all solid state amps (in contrast to tube amps) are designed to provide an output voltage which does not vary significantly as a function of load impedance, as long as the amp is operated within the limits of its voltage, current, power, and thermal capabilities.

Therefore, as a consequence of Ohm's Law and the definition of power, while of course not all solid state amps will double their rated MAXIMUM power capabilities into halved load impedances, nearly all of them WILL double the amount of power that is delivered into a halved load impedance (albeit perhaps with some increase in distortion), as long as they are operated within those maximum limits.

So when Ralph says:

If you have a single woofer and put one watt into it, it might make 92 db at one meter. If you have two of the same woofers in either parallel or in series and put 1 watt into the array, the result will be the each driver will have 1/2 watt on them- and the output will be exactly the same 92 db.

... Note that he refers to an input of 1 watt.

If he had referred to an input of a given number of volts, say 2.83 volts (which corresponds to 1 watt into 8 ohms, as he indicated), applying that voltage to two identical woofers in parallel (which together have half the impedance of a single woofer) would cause twice as much power to be delivered as if the same voltage were applied to only one woofer. Resulting (everything else being equal) in a 3 db increase in SPL. And conversely, applying that voltage across a combination of two identical woofers in series (which together have twice the impedance of a single woofer), would result in half as much power being delivered as if the same voltage were applied to only one woofer, resulting in a 3 db reduction in SPL.

Basically, the speaker designers Ralph referred to are thinking in voltage terms, rather than power terms. Which is arguably reasonable in the context of most solid state amplifiers, but can be very misleading when it comes to tube amps.

ZD, thanks for the nice words.

Best regards,
-- Al