Issue 223   May / June 2012
The Single-Ended Triode Paradox
Editorial By Robert Harley

   

"One of the worst-kept secrets in audio engineering is that what 
we hear does not always correlate with what we measure."
—Richard Heyser

"Whenever connoisseurship is found operating within science or technology we may assume that it persists only because it has not been possible to replace it by a measurable grading." — Michael Polanyi, Personal Knowledge: Towards a Post-Critical Philosophy

 

There's no greater paradox in audio than that of the single-ended triode (SET) amplifier. Ask an engineer to describe an amplifier's ideal technical characteristics and he'll describe everything an SET is not: low distortion and noise, high output power, high damping factor, low output impedance, ability to deliver current to low-impedance loads, generous dynamic headroom, wide bandwidth — the list could go on. So how can some of these amplifiers that are the antithesis of good textbook engineering sound so magical?

The SET's very existence calls into question fundamental beliefs and assumptions we routinely make about technical performance, sound quality, and the correlation between them. These amplifiers expose a crack in the edifice of audio engineering theory that is based on the conviction an amplifier can be judged by its technical specifications or measured performance. To the "meter readers," who reject the listening experience and instead believe that nothing more can be known about an audio component beyond its technical performance, a modern amplifier with 18W of output power at 3% distortion is a joke. But put that 18W SET in a system with an appropriate-load loudspeaker and the sound will melt the heart of even the most hardened objectivist.

This paradox arises because the technical measurements that attempt to quantify amplifier performance are simply inadequate and incomplete. Each element of an amplifier's performance description (bandwidth, THD, for examples) is a static two-dimensional sliver of information that, when combined with all the other limited two-dimensional slivers (noise, damping factor, TIM, etc.), attempts to completely characterize what is in fact a highly complex, multi-faceted, three-dimensional dynamic system. In addition, that complex system interacts with a vastly more complex system — the human brain. Predicting an amplifier's sound quality, or judging it to be good or bad, based on existing criteria is like looking at a few still images from a movie and then attempting to discern from those static photos the movie's plot, characterizations, dramatic arc, and meaning.

The SET exposes the fact that certain aspects of amplifier performance are not quantified by the traditional measurement arsenal. In what data-set do we find an amplifier's directness of expression — that feeling of the musicians being present in the room making music contemporaneously? What spec tells us if the amplifier reproduces timbre with such realism that we get goosebumps? What graph plots musicality along one of its axes? The vexing problem is not so much that traditional measurements fail to describe these qualities. Rather, the real quandary is that an SET, an amplifier that exhibits such grossly flawed performance according to established criterion for "goodness," can sound so wonderful. How can an amplifier that is so "wrong" sound so right?

Some will suggest that listeners are merely responding to the SET euphonic distortion — that the SET sounds good because of its distortion rather than in spite of it. There's no question that the largely second-harmonic distortion component of an SET is much more sonically benign than the upper-order distortion components of Class AB solid-state amplifiers. But a first-rate SET amplifier's magical qualities go far beyond this simplistic interpretation. The SET's resolution of inner detail that, singularly, conjures up a strikingly vivid picture of the instrument creating the sound is certainly not merely a euphonic second-harmonic distortion artifact.

This essay is neither a renunciation of all amplifiers other than SETs nor an evangelical campaign for the world to embrace the single-ended triode amplifier. They are limited in the loudspeakers they can drive, exhibit other practical drawbacks, and are certainly not for every listener. Moreover, only a very few of them are uncolored enough to be used as a reference. But when I listen to music through a pair of Lamm ML2.2s, I can't help but question whether nearly a century of conventional wisdom about what makes an amplifier "good" has led us down the wrong path.

 

 

 

 

SUBSCRIBE!
Click here to subscribe to The Abso!ute Sound.