Red Shift: Doppler distortion in loudspeakers Doppler and DiAural

Sidebar: Doppler and DiAural

Although the initial fuss that accompanied Ray Kimber's 1999 launch of DiAural—the proprietary speaker-crossover technology developed by Eric Alexander—has died down, the fact remains that it was and still is promoted as a means of canceling the Doppler distortion introduced by microphones: "Doppler Decoding," in DiAuralspeak. To my knowledge, however, this claim has never been challenged in the audio press. Although DiAural's two patents (US 6,115,475 and 6,310,959 B1, copies of which you can download from the US Patent Office make no mention of Doppler distortion, it has been a central plank of the company's marketing message—as reported, for instance, in Barry Willis's Web news reports on February 22, 1999 and April 4, 1999), and in Ken Kessler's writeup in the June 1999 Hi-Fi News.

How credible is the claim that a loudspeaker, whatever its crossover configuration, can cancel microphone Doppler distortion?

The first problem with this notion is that moving-coil loudspeaker diaphragms are mass-controlled through their working range, whereas high-quality (ie, capacitor) microphones are stiffness-controlled. This results in quite different behavior in respect to diaphragm displacement vs frequency. Whereas, assuming constant sound-pressure level, the loudspeaker diaphragm's displacement decreases with the square of frequency (at least, until it starts to become significantly directional), the microphone diaphragm's displacement is constant. It follows that, as a result, their Doppler distortions vary quite differently with frequency. To further complicate matters, dynamic microphones are resistance-controlled and ribbon microphones are mass-controlled, so diaphragm displacement vs frequency behavior is different for different microphone types.

Second, for Doppler Decoding to work even notionally, it would have to rely on the recorded and reproduced SPLs being matched, and on absolute polarity being maintained throughout the recording/reproduction chain. Obviously, these conditions are unlikely to be met, particularly in the case of multimiked or multitracked recordings.

Third, for a microphone to introduce significant levels of Doppler distortion, its diaphragm would have to undergo large excursions—of the same order as the loudspeaker diaphragm, if cancellation is to be feasible. In fact, microphone diaphragm excursions are minuscule. I asked Stephan Peus, president of development at Georg Neumann GmbH, to provide me with some representative figures. He e-mailed back a document, "Some Amazing Facts with Condenser Microphone Capsules," which quotes the diaphragm excursion for Neumann's KM 184 microphone (a miniature cardioid) as being just 10 nanometers—that's four-tenths of a millionth of an inch, or about a 40th the wavelength of blue light—at an SPL of 94dB. Compare this with the 7.5mm peak excursion required to generate this SPL at 100Hz in free space and at 3m (10') listening distance, using a drive-unit of 200mm (8") effective diaphragm diameter. The two figures differ by a factor of 750,000! Depending on their mechanical characteristics, the diaphragms of other capacitor microphones may undergo larger excursions, but their displacements will still be orders of magnitude smaller than a loudspeaker cone's.—Keith Howard