Microscopic Microphones

The job of a sonic transducer is simple: to convert an audio signal from one form into another. Speakers convert electrical signals into acoustic sound

The job of a sonic transducer is simple: to convert an audio signal from one form into another. Speakers convert electrical signals into acoustic sound waves by vibrating a diaphragm in response to an alternating current. Similarly, microphones convert acoustic sound waves into an electrical signal when a diaphragm vibrates in response to fluctuating air pressure.

The design of these transducers has not changed much since their development decades ago. For example, most microphones remain relatively large, bulky devices to house the associated electronics and protect the delicate diaphragm from damage. However, there are several new microphone designs that could change all that.

Among the most interesting approaches to microphone miniaturization is the Microflown. Invented by Hans-Elias de Bree in 1994, the Microflown does not measure fluctuating air pressure; instead, it measures the velocity of air particles across two tiny, resistive strips of platinum on silicon nitride, which are heated to about 200 degrees Celsius. (In fluid dynamics, the motion of gas or liquid particles is called flow, hence the name Microflown.)

When air flows across the strips, heat is carried from one to the other, causing a temperature differential between them. This results in a resistive differential, which generates a minuscule voltage proportional to the particle velocity. In the case of a sound wave, the air flow across the strips alternates according to the waveform, which results in a corresponding alternating voltage.

It's important to understand that an air particle is not a gas molecule but rather a volume of air that is small compared with the dimensions of the measuring device (which I'll discuss in a moment) or the sound's minimum wavelength (about half an inch at 20 kHz in air). The amplitude of the sound wave determines the particle velocity; for example, 94 dB SPL corresponds to a particle velocity of 2.5 millimeters per second, and 0 dB SPL (the threshold of hearing) corresponds to a particle velocity of 50 nanometers per second. The particle velocity can be increased at the sensors by placing carefully designed obstacles very nearby. These obstacles can form the structure in which the sensors are housed, which can increase the amplitude of the alternating flow by as much as 20 dB.

The entire Microflown (sensors and electrical connections) measures a mere 3 mm wide, 2 mm long, and 0.3 mm thick (see Fig. 1). The sensor strips are an amazing 0.0003 mm thick, making them impossible to see with the naked eye. In addition, the Microflown has no moving parts, which means it's highly reliable and exhibits no resonances. It's also more resistant to extreme ambient conditions, such as moisture and dirt.

To develop the Microflown and its various commercial applications, de Bree started Microflown Technologies (www .microflown.com) in 1998 with Alex Koers. Microflowns are fabricated in a clean-room environment using techniques similar to those found in semiconductor manufacturing. As a result, their performance characteristics fall within tight limits; for example, differences in sensitivity between individual units should be less than 1 dB, which is quite low compared with traditional microphones.

The Microflown is highly directional (the polar pattern is figure-8 all the way down to 0 Hz), and because it measures air velocity instead of pressure, it is very sensitive to near-field sources while effectively rejecting far-field sources. Its frequency response extends from 0 Hz to beyond 20 kHz, but its natural sensitivity drops by 6 dB/ octave above 1 kHz. This can be corrected with the appropriate electronics, and a low-noise studio mic is planned for next year. With companies like Sennheiser and Bruel & Kjaer expressing serious interest in Microflowns, who knows what exciting products might emerge in the future? A