Home |
Search |
Today's Posts |
#1
|
|||
|
|||
KISS 116 by Andre Jute
KISS 116 by Andre Jute
This text is copyright Andre Jute 1997, 2004 and may not be reproduced except in the thread KISS xxx on rec.audio.tubes THE VOLTAGES IN THIS AMP WILL KILL YOU. GET EXPERIENCED SUPERVISION IF IT IS YOUR FIRST TUBE AMP. Is distortion in SE amps euphonious? André Jute There are some who claim that distortion in single-ended amps don't matter, and even some who claim that their distortion signature is what creates the euphony of SE! The argument goes that what has gone wrong with solid state amps is that the search for ever lower total harmonic distortion is the wrong quest. To jump from there to SE distortion being a good thing is a world record in logical long jumps. In 1947 H. F. Olson did a series of amplifier distortion listening tests, most accessibly summarized on p609 of the Radiotron Designer's Handbook, 4th edition. The finding that attracts attention from the proponents of the 'euphonious distortion' camp is that a number of critical listeners to a class A SE amp with an HF bandwidth of 15kHz judged perceptible total harmonic distortion to begin at 0.75%, with tolerable distortion at 1.8% and distortion being noted as objectionable at 2.5%. At the time, this may have proven that below x% distortion did not matter too much. But it seems to me that fifty years later further consideration is required even if only because psychological research, unlike the physical sciences, does not pretend to offer definitive permanent answers. First, toleration of THD decreases with increasing bandwidth. With the complete human hearing range of 20Hz-20kHz now being the aspiration of all serious SE amps, if the Olson tests were repeated with a modern SE tube amp the results would, I believe, report lower tolerances for that reason alone. This is merely the headline: the case is complicated by the fact that with reduced bandwidth objectionable distortion can be disproportionally higher than perceptible distortion in relation to the tolerances at greater bandwidths. Second, tolerance for distortion is related to experience; this was well understood even in Olson's day. Those 'critical listeners' of 1947 were likely to have had less experience of very low THD reproduction chains and recordings than a similar group sixty years later. Clearly, today everyone interested has much greater exposure to high quality reproduction and also to live concerts. It seems at least possible that a similar group of 'critical listeners' selected today to take part in an equivalent test would establish lower tolerances for each parameter with a ceiling of 15kHz (and much lower tolerances at 20kHz). Thus far the case as it stands against the 'euphonious distortion' argument. There are however also two serious arguments that tend to support it: Third, higher toleration for distortion on any parameter is reported by Olson for SE amps than for PP. Fourth, modern SE tube amps have THD which compared to modern SS amps puts them in the neanderthal era of audio, even if vastly improved when compared to the classic designs. Despite all this, SE amps continue to be highly valued by significant numbers of 'critical listeners'. There is also a less incontrovertible point that I would like to add from personal experience and talking to designers and DIYers: it is possible by fitting the wrong combination of modern parts to refine any amp, but particularly an SE amp, to a point where it sounds sterile, at which point a common response from experienced listeners is, 'It sounds just like a tranny amp.' (As I say, this is not an uncontroversial point. There are many respected designers who believe that any propertly designed amp should sound like any other properly designed amp, regardless of its construction with transistors or tubes. They proceed from the assumption that the purpose of an amplifier is to reproduce the music as it was played: they believe that the aim of hi-fi is ever higher fidelity in an ever closer approach to the live performance. Others, having given up on that dream, consider that the aim of hi-fi should be to tailor a reproduction chain to their personal taste. Of course parties who start from such radically opposed basic assumptions will not agree on anything.) Thus, if I am wrong on the second point, increased experience, and a new set of Olson tests report approximately the same tolerances in a by now much more experienced group of critical listeners, the 'euphonious distortion in SE amps' argument would receive a shot in the arm at least to the extent that there would be corollary evidence that distortion up to a certain level adds to the perceived sonic advantage of SE over PP. (The modern critical listener will then be seen to have traded off an irrelevant potential reduction in noise for greater usable power. At that point we will then have the fascinating new question to answer: what inspired that minority of well-regarded critical listeners to reduce noise by choosing higher impedance output transformers, thus opening the whole can of worms again!) If my hypothesis is correct, and in a new study lower percentages are reported at all Olson parameters, the euphonious distortion argument will be exposed as a relative environmental factor - - with a further lessening of its importance being foreseen over time. Meanwhile, there is not enough evidence to take sides. © 1997, 2004 André Jute |
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
KISS 117 by Andre Jute | Vacuum Tubes | |||
KISS 115 by Andre Jute | Vacuum Tubes | |||
KISS 114 by Andre Jute | Vacuum Tubes | |||
KISS 113 by Andre Jute | Vacuum Tubes | |||
KISS 102 by Andre Jute | Vacuum Tubes |