A microphone is a straightforward device — sound goes in, voltage comes out. So shouldn't reading microphone specifications be a straightforward process? Mic specs are intended to tell you at a glance what to expect from a given microphone — specifically, how analogous the voltage is to the sound source.
FIG. 1: This figure shows three common microphone polar patterns. An omnidirectional mic (left) responds uniformly regardless of angle of incidence, a cardioid mic (middle) rejects sound from behind, and a bidirectional mic (right) rejects sound from the sides.
Unfortunately, however, by the time you've researched a microphone and waded through the various acronyms, such as EIN, mV/Pa, and dB SPL, your head hurts too much to make a rational buying decision. The truth is that no amount of research into microphone specifications can tell you what a mic will really sound like. Specs may help you narrow the field, but they are no substitute for your ears.
Once you know the terminology, though, mic specs can be helpful. In this article, I'll explain the concepts and terms you need to know to get the most from spec charts.
Aside from transducer type — dynamic, condenser, or ribbon (see “Square One: A Change Is Gonna Come” in the November 2006 issue of EM, available at www.emusician.com) — the first thing to look for in a microphone is its polar pattern, or directional response (see Fig. 1). A circular graph is used to illustrate the microphone's polar pattern, showing where in a 360-degree radius the mic is most sensitive. Some microphones offer multiple polar patterns, enhancing their usefulness.
A microphone with uniform sensitivity in every direction and plotted as a circle on the polar chart is said to be omnidirectional. Mics that are most sensitive to sounds arriving from the front, filling half (180 degrees) of the circular polar chart, are called directional. Bidirectional mics, on the other hand, are sensitive to sounds coming from the front and rear, but not the sides. The term on-axis is used to describe sounds that arrive at the sensitive parts of the pickup pattern. Off-axis refers to sounds hitting the areas of a mic's pickup pattern that are least sensitive.
Cardioid, so named because it resembles a heart shape, is the most common directional mic pattern, while figure-8 is the classic bidirectional pattern. There are a number of variations, such as wide cardioid (which is a compromise between cardioid and omni), that fall in between the basic pattern types. Between cardioid and bidirectional, the pattern gets narrower toward the front of the mic, rejecting more sounds from the sides, while a bubble of sensitivity opens at 180 degrees to the rear. As you move from cardioid to supercardioid to hypercardioid, the pattern narrows and extends, until it eventually becomes bidirectional.
Microphones respond to some frequencies better than others. The broken lines in Fig. 1 demonstrate how polar patterns vary depending on the frequency range. Manufacturers also create a frequency-response graph that indicates the mic's response to a tone swept from 20 Hz to 20 kHz. In Fig. 2, the tested microphone is about 3 dB more sensitive between 3 and 7 kHz than at 1 kHz — the frequency to which the curve is normalized.
FIG. 2: This figure illustrates a microphone''s typical frequency-response curve. The microphone enhances frequencies between 3 and 7 kHz. The boost below 500 Hz is due to the proximity effect.
Fig. 2 also shows two broken lines below 500 Hz, reflecting a property of directional mics known as the proximity effect. When a sound source is very close to a directional mic, the mic's low-frequency response is enhanced. This is the trade secret behind that rich, booming “announcer's voice” sound. The more directional the mic, the more pronounced the boost. The two broken lines reflect the mic's low-end response at a distance of 6 inches and 2 feet from the sound source, while the solid line indicates its response at 3 feet.
When you see a mic's frequency response listed as a numeric range, look for the degree of deviation — “±3 dB,” for example. Without that qualifier, the given frequency range is not helpful. Look carefully, and you'll see mics occasionally listed with a “20 Hz to 20 kHz” response that turns out to be 20 Hz to 15 kHz ±3 dB, with a significant rolloff toward 20 kHz.
If you will be working with very quiet sources, such as those in classical music or field recording, you will need a microphone that doesn't get in the way sonically. All mics have some degree of inherent noise, or self-noise. Self-noise is determined by comparing the inherent noise level with a hypothetical source sound whose level produces the same response from the mic (the equivalent input noise, or EIN). This is ordinarily measured in dB SPL, sometimes listed as “dB re 20 µPa,” and is often A-weighted. A-weighting approximates the nonlinear frequency response of our ears but also results in a better-looking (lower) number. If the figure is not A-weighted, it will refer to a specification known as CCIR 468-3, and the number will be about 10 dB higher. A mic with an EIN of less than 15 dB SPL A-weighted is very quiet.
You should also know about a microphone's sensitivity, the voltage it produces in response to an acoustic stimulus. A sensitivity of 70 mV/Pa means the mic produces an output of 70 mV when presented with an input of 1 pascal (94 dB SPL). Sensitivity may also be expressed in dBV/Pa. For example, a sensitivity of -37 dBV/Pa indicates a microphone whose output is 37 dB lower than that of a hypothetical microphone that produces 1V in response to 1 pascal. Converting from one scale to the other requires a bit of math, but in general a range of 10 to 100 mV/Pa translates to an approximate range of -40 to -20 dBV/Pa.
To achieve optimum performance from most modern microphones, the microphone preamplifier's input impedance should be much higher than the output impedance of the microphone itself. Some mic manufacturers recommend a ratio of 5:1, while others recommend 10:1. This type of high-to-low ratio is called impedance bridging, and it results in maximum transfer of voltage from mic to preamp in order to minimize signal degradation.
A mic's spec sheet will list the mic's output impedance. If it doesn't also list a recommended load impedance for the preamp, you can simply multiply the output impedance by 5 or 10, depending on which of the above ratios you want to work with.
With a transformer, a signal passes through a coil of wire wrapped around a magnet and induces signal in an adjacent coil/magnet assembly, allowing signal to pass without physical contact between the two assemblies. Changing the number of turns of wire in the output coil, or winding, changes its impedance relative to the input winding, thereby changing the output signal level.
In general, a mic with an output transformer allows you to use a longer cable without fear of unacceptable signal loss, and it also increases resistance to noise. Intense low-frequency sounds, however, can overload a transformer, causing distortion. Some engineers will therefore prefer a transformerless design if they can get by with relatively short cable runs, as in a studio (as opposed to a live setting, in which long cable runs are common).
There are some international standards that attempt to codify the way in which microphones are tested and their specifications reported, but few manufacturers follow them faithfully. A wise mic shopper therefore needs to come armed with a broad and flexible vocabulary, a probing mind, and a healthy dose of caution, if not outright skepticism. So armed, one can glean much useful information from Web sites, spec sheets, and glossy sales literature.
Of course, all of that research only serves to narrow the field down to which mics are likely to please your ears. After that, you must judge which mic sounds best on which source.
Brian Smithers is a musician, composer, and engineer in Orlando, Florida. He teaches at Full Sail Real World Education and Stetson University.