Need help understanding tube wpc

"After all, musical peaks are voltage driven, not current driven."

In the world I live in, musical peaks being reproduced by a power amp produce power, not merely voltage. Voltage in a speaker cannot exist without current and this is inviably defined by Ohm's Law and the Power formula (Power=Voltage X Current).

So- a musical peak delivered by a power amp has a voltage component- *and* a current component. Since the peak represents a peak in power as well as voltage, current must therefore be peaked also.

The idea that the peak is voltage driven comes from the Voltage paradigm I've mentioned in a few threads here. What I've not mentioned is that paradigm is actually that- a paradigm, and not one based on reality. It is in fact an artifact of the 50s and 60s when transistors were making their way into audio. A central precept of the Voltage paradigm is the use of negative feedback (ostensibly to reduce THD); such use is in violation of the rules of human hearing which we all subscribe to by default.

The violation, for those curious, is the addition of odd-ordered harmonic content that negative feedback brings.

A curious artifact of the Voltage paradigm is that nothing but voltage matters in the response of an amplifier. Another is that 'voltage source' amplifiers are also defined as 'high current'. Yet another is the idea that speakers are 'voltage driven'. English speaking people will note some contradictions.

The Power paradigm aims to correct these oddities. First, *all* speakers are power driven. All power amplifiers produce power. 'High current' does not exist for musical reproduction without the generation of power, and the same is true of voltage. In this way, the power formula and Ohm's Law are satisfied within the conversation of power amplifiers and speakers and at the same time the meanings of English words are also satisfied.

Within the conversation of this thread, the issues relating to why tubes are somehow able to produce more *usable* power when their total power is less than that of transistors is easily revealed by the Power paradigm, which has it roots based on the rules of human hearing, rather than a thought model conceived to sell transistor amplifiers in particular.

The answer is that tubes generate power in a way that satisfies more of the rules of human hearing than transistors do. For example, SETs get their dynamic punch out of their harmonic generation: odd ordered harmonics are masked by even orders, so while the amp sounds lush, the odd orders are triggering the human ear to hear dynamics on peaks. It is an illusion.

Other, lower distortion tube amplifiers still manifest a greater percentage of *usable* (musical) power by the simple use of components that are inherently more linear than transistors, and usually with less stages (meaning less places for things to get messed up), objectionable distortions are minimal, resulting in little or no feedback being required for the amp to do its job. Thus the human ear is not *as* offended, and the bottom line is more of the amplifier's power generated can be used for meaningful musical reproduction.

In the Power Paradigm the rules of human hearing are not ignored so an amplifier can measure well and sound good too because the pertinent specifications that are important are the ones that get measured.

In the Voltage paradigm, as Pauly points out:

"I have never found any correlation between measured numbers like THD and what the amp sounds like to my ears. In fact have heard amp with excellent THD number that are useful only as doorstops."

-that there is no correlation between specs and sound. There is a huge disconnect here! The Voltage Paradigm seems to equate to the Emperer's new Clothes. After nearly 50 years- *that* would seem to be a little old :)

This is simple Ohm's Law. One can try to make it as complicated as they like, but it is actually just as simple as it sounds.

The amount of current needed in the circuit is dictated by the impedance of the load at the frequency that is being reproduced. After that, it's all a matter of whether or not the amplifier can swing enough voltage in time to keep up with the amplitude of the signal as it changes on a dynamic basis.

Since we've already acknowledged that high current capacity is not as necessary as most would think with reasonable speaker loads, the only reason that an amp would clip has to do with running up against a limited voltage capacity and / or the inability of the circuitry to slew fast enough to deliver the voltage required. Both result in clipping, whether it is due to limited headroom and / or a lack of speed.

With high rail voltages, high current capacity, high speed / wide bandwidth and good circuit stability, you can drive any load that you want with reasonably low levels of distortion. If you limit even just one variable in the aforementioned list though, the versatility and performance potential of the circuit is drastically reduced. This dictates more careful matching of the associated load ( speaker and speaker cables ) to that of the performance limited amplifier.

Since all amplifiers are limited in one way or the other to some degree, and no speaker is purely resistive with a higher nominal impedance, some matching is always required. Having said that, the more competently designed the amp, the more consistent the performance that it will deliver, regardless of the load. Obviously, one can go to extremes coming up with radical speaker loads that might embarrass all but the most advanced amplifier circuitry, but as a general rule, these are just that i.e. extremes and not the norm.

If speaking of extremes, anything is possible and specifics must be mentioned if we are to have any type of meaningful conversation. I have strictly been speaking in general terms, as i've not seen any mention of specifics in this thread. Sean
>

PS... The fact that Ralph and i are "debating" should tell you folks quite a bit. That is, even though we both find many of same attributes very desirable i.e. wide bandwidth, high speed, reasonably low distortion, low negative feedback, high stability, high bias circuits, etc... there is still plenty of other things for us to disagree about. This is one of the reasons there are SOOOOO many different products out there with different designs.

As can be seen by our responses here, most of this boils down to what the best way is to achieve all of those goals simultaneously without having to cut a lot of corners to get there.

Not to change the subject, but i wanted to point something out. Ralph aka Atmasphere said: "Since the peak represents a peak in power as well as voltage, current must therefore be peaked also".

This is not always true. When a speaker resonates, it produces a peak in output, even though voltage and current input from the amp are at a minimum due to the high impedance encountered. How intense this peak is has to do with the alignment of the speaker system and how well the amp can control the reflected EMF.

Some speaker designs seek to tame this form of "self oscillation" ( low Q sealed designs, transmission lines, etc... ) whereas others try to exploit it ( many vented designs ) in hopes of getting "free output". That "free output" doesn't really come "free" though, as this type of system typically lacks damping while producing rather inarticulate bass at or near the point of resonance(s).

Whereas "uncontrolled oscillation" is typically frowned upon anywhere near the audible bandwidth in every type of component that i can think of, this seems to be the accepted norm with many vented loudspeaker designs. This has always baffled me ever since i understood the sonic and technical differences in various bass alignments. Sean
>

In AC circuits there is often an inductive nature that can result in current lagging behind voltage. So it is true that current may not be peaked in all cases of loudspeaker; the more 'resistive' the speaker (i.e. magnetic planars, Avalon) the more the current will be peaked with the voltage.

Nelson Pass wrote a very interesting article in Audio Express about 2 years ago that offers some interesting insight into the issue of 'high current'. I agree with his conclusions, although it would be easier to understand if his language were not part of the Voltage Paradigm conversation (although for those who only understand the Voltage paradigm, his text is a nice introduction to some very real issues!). His more recent designs are clearly based on Power paradigm concepts.

I agree that Sean and I probably do have similar concerns regarding amplification! -despite the fact that we (apparently) have very different ways of going about it.

I prefer an amplifier with both wide bandwidth (I play bass so I need good 20Hz square wave response to be convinced) and low distortion if I can get it without putting a sheen or hardness up top. Having done a lot of live recordings over the years I have to say I've gotten picky and have yet to hear an amp with negative feedback that sounds right (although there was a time when I thought all that stuff was just fine- my personal victory over the past is to have found the breakthrough results that left that behind).

I do recommend to any audiophile who is trying to create a reference-quality system to start with a set of condenser microphones, an Ampex tube recorder (I use Neumann U-67s and an updated Ampex 351-2) and some living musicians. Create a library of recordings in different venues; very qucikly you find out what a large variance exists between tube and transistor playback!! Its great to spend time at live concerts but you are powerfully in touch with what is happening when you also have decent recordings of them!

I hope y'all had a great Thanksgiving!!