Music Power
In their first flush of enthusiasm for solid-state audio amplifiers, manufacturers and advertising copy writers collectively made the happy discovery that most inexpensive audio amplifiers powered by simple supply circuits, such as that shown in Figure 5.1(b), would give a higher power output for short bursts of output signal, such as might quite reasonably be expected to arise in the reproduction of music, than they could give on a continuous sine-wave output. This short-duration, higher output power capability was therefore termed the music power rating, and, if based on a test in which perhaps only one channel was driven for a period of 100 ms every second, would allow a music power rating to be claimed that was double that of the power given on a continuous tone test in which both channels are driven simultaneously (the so-called rms output power rating).
Influence of Signal Type on Power Supply Design
Although this particular method of specification enhancement is no longer widely used, its echoes linger on in relation to modern expectations for the performance of hi-fi equipment. The reason for this is that in the earlier years of recorded music reproduction there were no such things as pop groups, and most of those interested in improving the quality of recording and replay systems were people such as Peter Walker of Quad or Gerald Briggs of Wharfedale Loudspeakers, whose spare-time musical activities were as an orchestral flautist and a concert pianist and whose interests, understandably, were almost exclusively concerned with the reproduction, as accurately as possible, of classical music. Consequently, when improvements in reproduction were attempted, they were in ways that helped enhance the perceived fidelity in the reproduction of classical music and the accuracy in the rendition of the tone of orchestral instruments. In general, this was easier to achieve if the electronic circuitry was fed from one or more accurately stabilized power supply sources, although this would nearly always mean that such power supplies would have, for reasons of circuit protection, a fixed maximum current output. While this would mean that the peak power and the rms power ratings would be the same, it also meant that there would be no reserve of power for sudden high-level signal demands—a penalty that the tonal purists were prepared to accept as a simple fact of life.
However, times change and hi-fi equipment has become easier to accommodate, less expensive in relative terms, and much more widely available. Also, there has been a considerable growth in the purchasing power of those within the relatively youthful age bracket, most of whose musical interests lie in the various forms of pop music— preferably performed at high signal levels—and it is for this large and relatively affluent group that most of the hi-fi magazines seek to cater.
The ways in which these popular musical preferences influence the design of audio amplifiers and their power supplies relate, in large measure, to the peak short-term output current that is available since one of the major instruments in any pop ensemble will be a string bass, whose sonic impact and attack will depend on the ability of the amplifier and power supply to drive large amounts of current into the LS load, and it must do this without causing any significant increase in the ripple on the DC supply lines or any loss of amplifier performance due to this cause.
A further important feature for the average listener to a typical pop ensemble is the performance of the lead vocalist, commonly a woman, the clarity of whose lyric must not be impaired by the high background signal level generated by the rest of the group.
Indeed, with much pop music, with electronically enhanced instruments, the sound of the vocalist, although also enhanced electronically, is the nearest the listener will get to a recognizable reference sound. This clarity of the vocal line demands both low intermodulation distortion levels and a complete absence of peak-level clipping.
The designer of an amplifier that is intended to appeal to the pop music market must therefore ensure that the equipment can provide very large short-duration bursts of power; that the power supply line ripple level, at high output powers, must not cause problems to the amplifier; and that, when the amplifier is driven into overload, it copes gracefully with this condition. The use of large amounts of NFB, which causes hard clipping on overload, is thought to be undesirable. Similarly, the effects of electronic (i.e., fast acting) output transistor current limiting circuitry (used very widely in earlier transistor audio amplifiers) would be quite unacceptable for most pop music applications so alternative approaches, mainly based on more robust output transistors, must be used instead.
In view of the normal lack in much pop music of any identifiable reference sound source, such as would be provided by the orchestral or acoustic keyboard instruments in classical music forms, a variety of descriptive terms has emerged to indicate the success or otherwise of the amplifier system in providing attractive reproduction of the music. Terms such as “exciting,” “giving precise image location,” “vivid presence,” “having full sound staging,” “blurred,” or transparent are colorful and widely used in performance reviews, but they do not help the engineer in his attempts to approach more closely to an ideal system performance—attempts that must rely on engineering intuition and trial and error.