How Does a Microphone Work? Turning Air Pressure Into Voltage

Almost every microphone in use is one of two kinds, and the choice between them is usually made on character and robustness rather than on specification:

• •Dynamic microphones, in which a coil attached to the diaphragm sits in a magnetic field, so movement induces a voltage by exactly the same principle as a generator; they need no power, tolerate enormous sound levels, survive being dropped and handle heat and damp, at the cost of a slower response to fine detail because the coil has mass
• •Condenser microphones, in which the diaphragm forms one plate of a capacitor with a fixed backplate, so movement changes the capacitance and therefore the voltage; the moving part can be extremely light, which gives excellent sensitivity and high frequency detail, and the design requires a polarising voltage and a built-in preamplifier, supplied as phantom power down the same cable
• •Ribbon microphones, a variant of the dynamic type using a thin corrugated metal ribbon in a magnetic field, prized for a smooth top end and historically fragile
• •Electret condensers, which hold a permanent charge in the material itself so no polarising voltage is needed, which is what almost every microphone in a phone, laptop or headset is
• •Microelectromechanical microphones, condensers etched into silicon, small and cheap enough to put several in one device for noise cancellation and direction finding

What a microphone picks up from different directions is described by its polar pattern, and the pattern comes from the physical construction rather than from any processing. A capsule sealed at the back responds only to the pressure at its front and therefore picks up sound equally from every direction, which is omnidirectional. Open the back so that sound reaches both sides of the diaphragm and the capsule responds to the difference in pressure between front and back, which cancels sound arriving from the side and produces a figure of eight pattern picking up front and rear equally and rejecting the sides. Combining the two behaviours in one capsule gives the cardioid pattern, sensitive at the front and rejecting the rear, which is the most common choice for speech and for isolating one source. Pressure gradient designs carry two side effects: they are far more sensitive to wind and breath, which is why they need windshields, and they show proximity effect, a rise in bass as the source gets very close, which announcers exploit deliberately and which makes a close-miked voice sound fuller.

Frequency response describes how evenly the microphone responds across the audible range, and a deliberately uneven response is frequently preferred, since a presence rise around a few kilohertz helps a voice cut through. Sensitivity says how much voltage a given sound pressure produces, which determines how much gain the preamplifier must add. Self noise, quoted as an equivalent sound level, sets the quietest thing that can be recorded usefully and matters enormously for quiet acoustic sources. Maximum sound pressure level sets the loudest, beyond which the signal distorts, which is why a condenser in front of a drum or guitar amplifier may need its pad switch. Impedance and the balanced connection matter for cable runs: a balanced line carries the signal twice, once inverted, so interference picked up along the way appears identically on both and cancels when they are recombined, which is why professional microphone cables can run long distances without humming.

The single largest influence on a recording is where the microphone is put, because it hears the room as well as the source. Moving closer increases the ratio of direct sound to reflected sound, giving a drier and more detailed result, and moving away picks up more of the space. Because sound reflects, a microphone frequently receives the same signal twice with a short delay, and when that happens the two copies reinforce at some frequencies and cancel at others, producing comb filtering, which sounds hollow and is the usual explanation for a recording that is inexplicably thin. The same effect arises when two microphones on one source are at different distances, which is why the three to one guideline exists, keeping microphones at least three times as far from each other as from their source. Reflections from a nearby hard surface, including a desk or a music stand, cause the same problem, which is why an acoustic treatment on the first reflection points changes a recording more than most equipment upgrades.

A microphone converts pressure changes into voltage by letting them move a light diaphragm. Dynamic types generate voltage in a moving coil, need no power and survive abuse; condensers vary a capacitance, need powering and hear more detail because the moving part can be lighter. Directionality comes from whether sound can reach the back of the diaphragm, which makes gradient microphones susceptible to wind and gives them proximity effect. Reflected sound arriving late produces comb filtering, which is why placement matters more than equipment.