Boundary placement boosts low-frequency output by 6 dB per solid-angle halving.
Evidence
- First, at the lowest frequency, the curves are separated by very close to 6 dB for each halving of the solid angle, as Figure 13.1 predicts. The fact that there is a tiny discrepancy can easily be accounted for by the fact that the large powerful woofer is not exactly a vanishingly small “point monopole”, it is not located “in” the wall/floor, and the building was sheathed in corrugated metal siding, which is not a perfect reflector. In any event, the approximate 6 dB difference begins to diminish immediately and just below 200 Hz all of the curves converge in a null. The reason is that the woofer was facing forward, with the rear of the enclosure against the wall. This separation allowed the sound to travel backwards from the driver to the wall, and back again toward the driver where it interacted with sound being radiated by the diaphragm. The round-trip distance, about 36 inches (914 mm), is one-half wavelength, at 188 Hz, which is where the destructive acoustical interference occurs. … The adjacent boundary phenomenon was known to acousticians (Waterhouse, 1958), but it was Allison and Berkovitz (1972) and Allison (1974) who brought it into the consciousness of audio people in papers that described, in measurements, the significant dimensions of the problem. It is sometimes known as the “Allison effect”. — Toole, Floyd E, Sound reproduction- the acoustics and…, 1459
Categorization
- DEFINE: The ‘Allison effect’ gets its named definition here: the 6 dB-per-8π-halving gain from adjacent boundaries, limited by quarter-wavelength spacing.
- QUESTION: Worth chasing: the exact interaction described — rear-face reflection arriving back at the diaphragm — is a specific mechanism that could generalize to other driver/enclosure geometries.
- XREF: connects to: Operating with three nearby boundaries gives a fixed 9 dB low-frequency rise — David Howard, Jamie Angus, coustics and Psychoacoustics, 2193