Solid State Preamp Reviews

Why should the Cascade Basic be so sensitive to polarity reversal? I can only speculate and point an accusing finger at the possibility that inherent phase-response errors are compounding the effects of polarity reversal. A nonlinear phase response will produce a frequency-dependent phase shift of harmonics in a complex tone, the end result being a distortion of the signal waveform. So the broad issue is whether the auditory system is sensitive to waveform distortion or to changes in waveform slope (a polarity reversal changes the slope of the signal; ie, a compression with positive slope is presented as a rarefaction with negative slope. By now there exists a massive body of scientific literature that supports the audibility of monaural phase effects (the technical name for this phenomenon), although the biological basis for it has not yet been pinned down.

Clark Johnsen, in his book The Wood Effect (see September 1988, p.5), paints Helmholtz as being partly responsible for the current indifference to absolute polarity because of the latter's influential view that the ear is "phase deaf." This is based on Helmholtz's famous dictum that "...differences in musical quality of tone depend solely upon the presence and strength of partial tones, and in no respect on the differences in phase under which these partials enter into combination." Thus, the ear/brain is supposed to act strictly as a spectrum analyzer without regard to waveform. In truth, Helmholtz did hedge his phase rule by stating that harmonics beyond the sixth to eighth do produce dissonances and beats that make a phase effect possible, at least for these higher harmonics.

I find it remarkable that anyone in the middle of the 19th century could study phase effects at all, and I think that Helmholtz's findings are quite consistent with the severe limitations of his non-electronic, purely mechanical apparatus. For Johnsen to imply that somehow Helmholz impeded the progress of perceptual acoustics is about as credible as to claim that Newton impeded the development of relativistic physics.

The next significant milestone in the study of phase was the work by Mathes and Miller at the Bell Telephone Laboratories in 1947. They found that the envelope-wave shape of a complex steady-state tone is an important factor in the perception of tone quality: roughness vs smoothness and a sensation of apparent pitch. As waveform shape depends on the phases as well as the amplitudes of the components, the above perceptual differences could be produced by changes in phase alone of even a single component or group of components.

In 1962, Craig and Jeffress reported work conducted in the footsteps of Charles Wood, who, in 1957, while at the Defense Research Laboratory, informally discovered the audibility of polarity inversion. They found that listeners could consistently assign different sensory qualities to different phase relations. However, a surprising finding was that certain combinations of phase and level in complex waveforms produced a tone which, when reversed in phase, sounded higher-pitched, louder, or purer to some listeners, and lower-pitched, softer, or less pure to others. Note that a phase reversal of the complex tone would reverse the subjects' judgments, but would still leave them in disagreement.

Hence, it would appear that there are significant individual differences in the perception of phase, and that the concept of "absolute polarity" seems to be in jeopardy. A polarity that improves the sound quality for me may produce the opposite result for someone else.

Another important caveat about observing phase effects in the "real world" has to do with the listening room. Typically, phase effect experiments are done with headphones so that room effects are eliminated. In a reverberant environment, strong room reflections will sufficiently alter the direct signal waveform to mask phase effects at the source: If your room is alive, you may have a difficult time discerning polarity reversal. This is another good reason to treat your room. You might even wish to defer purchasing that expensive speaker cable you've been lusting after in favor of wall-absorptive treatment.

Craig and Jeffress go on to offer several theoretical possibilities for explaining the ear/brain's phase sensitivity. There are indications that the waveform of a complex tone, by effecting small changes in the particle movement of the cochlear fluid, is able to shift the place of maximum excitation along the cochlear membrane, which would account for timbre changes due to phase. There is also some evidence that the steepness of the waveform affects the sensation level. The great Bekesy, for example, found while investigating beats between a squarewave and a sinusoid that a more intense sensation was associated with the steeper wavefront. Another possible mechanism involves phase-related changes in the temporal response of neural stimulation. The neural response in the basal turn of the cochlea may be strongly affected by the waveform and signal steepness.—Dick Olsher