[Synchrodyne]

I think that a significant element of the crossover distortion problem, namely the inherent asymmetry of the Lin quasi-complementary output stage, had been addressed, in terms of cause, effect and remedies, in the late 1960s by several circuit designers, including Bailey (WW 1968 May p.94ff), Shaw (WW 1969 June p.265ff) and Baxandall (WW 1969 September p.416ff.) Baxandall’s graphical explanation was quite clear. He reprised that work in 1977 (in Amos, Radio, TV & Audio Technical Reference Book), expanding it to show that the Quad triples (of 1967) achieved essentially the same result as his “Baxandall diode” of 1969. (Baxandall elsewhere claimed credit for suggesting the triples to Quad, but he did not claim to have originated the idea.)

In 1966, Bailey had noted the transistor amplifiers could be problematical in terms of listener fatigue, etc, and chose transformer drive as the best option at the time. His reasons for so doing were not covered in detail, though, until his write-up on his 1968 fully-complementary design, as mentioned above. In his 1966 article, (WW 1966 November p.542) he also addressed the germanium vs. silicon issue, including the comment: “In fact germanium transistors usually give far lower distortion due to their better linearity”. It would appear that the change from germanium to silicon, whilst generally beneficial, may have exacerbated the distortion issue.

With all of that work going on, with key points in the public domain, one might have expected equipment reviewers to have been well-informed as to the need to at least look at the known problem areas for transistor amplifiers. In fact my recollection is that some did, and for example Gordon J. King was on to the crossover distortion aspect in the late 1960s.

I also have the impression that some reviewers of the 1960s, such as Geoffrey Horn and John Borwick, made a lot more measurements than were actually published, and probably more than the commissioning magazines had paid for. Certainly, rereading Horn’s Gramophone review of the Radford STA25 left me with the impression that he had “poked and prodded” the machine to death in an effort to find fault. As well as on the technical side, these folk also had credibility in that they were music afficionados and regular concert goers, so were familiar the “original sound”, as it were.

Nonetheless, insofar as the reviewing industry generally might have missed the boat in fully observing, measuring, and commenting upon the key issues surrounding the valve-to- transistor transition, they might have helped open the door to what had become “audiophoolery”. Nonetheless, this was also the time when the hi-fi industry moved from niche to mass-market status. That alone would have been cause enough for the snake oil merchants to want a part of it. And the larger market more easily allowed the entry of manufacturers who had unusual or extreme views, both rational and otherwise (e.g. “no tone controls”). A bigger industry invited more magazines, and playing to the unusual and extremes (the tabloid approach as it were) was probably seen as the better business plan in terms of gaining sales.

Regarding the importance of the transient behaviour of amplifiers, not revealed by steady-state testing, it would appear that in a general sense it was known before its 1970s rediscovery. This piece at least slightly predates that time: “Good transient response. In addition to low phase and frequency distortion, other factors which are essential for the accurate reproduction of transient waveforms are the elimination of changes in effective gain due to current and voltage cut-off in any stages, the utmost care in the design of iron-cared components, and the reduction of the number of such components to a minimum. Changes in effective gain during "low-frequency" transients occur in amplifiers with output stages of the self-biased Class AB type, causing serious distortion which is not revealed by steady-state measurements. The transient causes the current in the output stage to rise, and this is followed, at a rate determined by the time-constant of the biasing network, by a rise in bias voltage which alters the effective gain of the amplifier.”


Cheers,