Over the course of the next five installments I will address the workings and use of microphones. This will include both technical and practical aspects such as transducer types, patterns, proximity effect, spacing, and frequency response as well as instrumental, vocal, ensemble, live and room miking techniques. For now, I’ll begin with the basics: What types of microphones are there? How do they work? What makes one sound so much different than another?
To begin with, I want to make one thing clear; since the microphone is where the acoustic energy of a musical performance is first translated into an electrical signal, its selection and use are of utmost importance. The quality, properties and placement of the microphone will affect the rest of the production process. Factors such as polar pattern, sensitivity, transient response, frequency response, transducer type, power handling, self-noise, materials and dimensionality all affect the sonic character of a microphone and inform its usage. Choosing the right mic makes everything else in the recording/mixing process that much easier to do well.
So… what’s your type?
Microphones are transducers because they change one form of energy to another (acoustic to electric). They are classified into three main types by the way in which they accomplish this transformation. The categories are: dynamic, ribbon, and condenser.
Dynamic microphones consist of a coil of wire that is attached to a moving diaphragm and suspended within a magnetic field. When the diaphragm moves in response to acoustic vibrations, the coil moves within the magnetic field. This creates an electrical signal analogous to the original sound wave. This type of transduction is similar in concept to a moving-coil speaker design, only in reverse. (In fact, headphones have sometimes been used as dynamic microphones in a pinch or for achieving cool lo-fi effects, and Recording one published a DIY project for making a kick drum mic from a cheap speaker!)
Because of their basic design, dynamic mics do not require a power source to operate and tend to be quite rugged. For similar reasons, they are also fairly inexpensive (ca. $15-$450). At the same time, dynamic microphones are not the most sonically accurate of microphone types. This is not to say that they are undesirable. In fact, it is precisely the coloration that they impart upon a sound which makes them desirable in many circumstances.
In a ribbon microphone design, a thin metal ribbon is suspended in a magnetic field. When the ribbon moves within the field (again in response to acoustic vibrations) an electrical signal is generated. This operating principal is quite similar to that of dynamic microphones, except here the ribbon acts as both the diaphragm and the coil.
Ribbon microphones are generally quite accurate and are often considered to exhibit a warm and desirable tone. Because of the thin metal foil material used for the rib bon, traditional designs were quite fragile. More recent developments in both materials and design have allowed some newer models to be much more robust. Ribbons require more amplification (gain) than other microphone types and are generally more expensive than dynamics (ca. $500-$3000).
Condenser microphones change acoustical energy into electrical by varying capacitance (the ability to store electrical energy). In this case, a charged diaphragm vibrates and changes its proximity to an equally charged backplate. The changing distance between the two varies the capacitance and, therefore, the voltage. The name comes from the fact that this two-plate element, known as a capacitor to electrical engineers, was once called a condenser. In the USA, the old name stuck when talking about microphones, but you’ll sometimes hear British engineers speak of “capacitor mics.”
Because the diaphragm and backplate must be charged, condenser microphones require a power source to work. For traditional condenser designs, this was supplied via a 48-volt signal from the microphone preamp, called phantom power. Some condensers have a battery compartment built into the handle, typically for 1.5V AA batteries.
A relatively recent breakthrough in materials design has led to the proliferation of electret condensers, which do not require phantom power to operate. Instead, these microphones incorporate a new material that is permanently charged, not unlike a magnet but with an electric charge rather than a magnetic field.
Many people are under the mistaken impression that electrets do not require any power. They do! Not for the permanently charged backplate and diaphragm but for the internal amplifier. The amount needed, however, is very small. Tiny batteries can be enclosed within the microphone to supply this charge, and they can last anywhere from several years to over a decade, depending upon type and use. Because of these factors, I have known engineers who have taken electrets in to be “fixed”, when all they needed was to have these batteries replaced. It is useful to note, however, that even with a dead battery, most electret mics will still function fine if supplied regular phantom power.
While high-quality traditional condenser models tend to be expensive (ca. $500-$6000+), recent designs and manufacturing techniques have created a wide range of inexpensive condensers as well (ca. $100-$600). Though a number of these microphones can sound very good, keep in mind that the price break usually comes with a reduction in a models options and/or the use of lower quality components and looser manufacturing tolerances. Many of these microphones will not have the longevity or consistency of more expensive models. Due in part to simpler circuitry, however, electrets do tend to be less expensive than traditional condensers.
Though some newer models are fairly hardy, all condensers should generally be considered fragile. A drop from waist-high can be a condenser’s demise.
Microphones differ in their sensitivity to sound arriving at the capsule from various angles. One mic may pick up sound equally as well from the front and the rear, while another may be much less sensitive to sound coming from behind. These directional sensitivities are known as polar patterns.
These patterns are often expressed using a 2-dimensional graph which details how sensitive a microphone is from all angles in a 360-degree circle around the capsule. The 0-degree reference runs perpendicular to the front of the diaphragm. Sound waves hitting the microphone from that angle are said to be on-axis. Sound coming from behind is said to be 180-degrees off axis, while sound from the right or left is 90 or 270-degrees off axis, respectively (from the microphones point of view).
While there are actually infinite variations on polar patterns, they are generally summed up into five categories representing the most common basic types. Pictograms are often used to represent these five patterns (look for examples in the picture viewer):
- Omnidirectional - Equal sensitivity in all directions. Used to get the best sense of the acoustic environment. Also, since it tends to have the most accurate frequency response of the patterns, it is often used for scientific measurements and full-frequency classical recordings.
- Cardioid (unidirectional) - Most sensitive to sound coming at the front, less sensitive to sound coming from the sides, least sensitive to sounds from the rear. Cardioid patterns are useful for helping to isolate the sound of one particular instrument (or other sound source) from others nearby. For example, a cardioid microphone on the snare of a full drum kit can be angled in such a fashion as to be least sensitive to the toms and/or hi-hats, but still have greatest sensitivity to the snare sound itself.
- Supercardioid - Similar to cardioid, except has some sensitivity to the rear and rejects sound most from 120º and 240º off axis.
- Hypercardioid - Similar to bidirectional (see below), except has slightly less sensitivity to the rear and rejects sound most from 105º and 255º off axis. Both hypercardioid and supercardioid are good for use when rejection from specific rearward angles is desired. This could be used to mike a tom between two cymbals. These are also often used in live situations where there are two stage monitors for the singer, one to either side. (Note that in casual use the terms hypercardioid and supercardioid are often used interchangeably-not entirely correct, maybe, but the two types do exhibit a strong similarity.
- Bidirectional (or Figure Eight) - Equally sensitive to sound coming from the front and back, less sensitive from the sides 90º and 270º). Great for picking up a blend of two singers facing each other, but rejecting sound from the sides. Also often used to pick up both direct and reflected sound, and in stereo miking applications: a figure-eight is used with a cardioid for mid/side miking, and two figure eights are used to create a Blumlein pattern. (More on these in TCRM 11)
Though microphones are classified as employing one or more of these patterns, there are subtle (and not so subtle) differences in the exact patterns of different microphones. For this reason, mics often come with a polar pattern response chart specific to that model (sometimes even to the exact unit). Some are even more specific, measuring the polar response at various frequencies and distances, as both of these can make a very large difference in coloration and usage. Any differences of frequency balance between the 0º reference and other angles is known as off-axis coloration.
Though rare, some microphones documentation includes a three-dimensional response graph. The simpler two-dimensional ones assume three-dimensional symmetry, which is not always true. Be that as it may, a thorough three-dimensional analysis, including numerous frequencies and distances, can actually be unwieldy. Ultimately, a two-dimensional graph with high, mid and low frequency plots at close and middle distances (2-10 cm and .6-1.0 m, respectively) is enough to give an idea of how the mic behaves at different angles. Your ears, experience, and experimentation must do the rest in actual studio use.
Axis of address
With the grills and windscreens that are often attached to a microphone capsule, it is sometimes difficult, if not impossible, to see the diaphragm itself. This would not be a concern, except that it is important to determine which way it is actually facing on directional mics. Some models, such as the famous SM58, are end addressed. This means by aiming the end (or tip) of the mic directly at the sound source will be on-axis with the diaphragm.
Others mics, like the well-known Neumann U87, are side addressed. One side of the grill is the on-axis side, but it’s not always obvious which side that is! If there is a company logo, that’s a good first guess at the on-axis side, but many modern designs confuse the issue with brand names and logos on both sides. When in doubt, put on headphones and move the mic (or move around the fixed mic) and listen for the louder sound; when all else fails, read the manual.
You're getting close…
With directional microphones (everything other than omni), there is a boost of the lower frequencies as the microphone is moved closer to the sound source. This is known as proximity effect. It is one way that DJs and announcers get that special “larger than life” vocal sound we have all become so familiar with.
If a close sound is desired, but not the extra bass, equalization (EQ) can be used to offset this, usually in the form of a highpass filter or a negative gain low shelf. Some microphones have high pass filters (low-cut) filters built in to help combat this effect as well as general low-end rumble. Of course this filtering can be done in other places such as the preamp, mixing board, or in software.
As previously mentioned, a microphone is not equally sensitive at all frequencies. For this reason, a graph of a microphones response over the audible 20 to 20,000 Hertz range is usually also included in its documentation. As with polar pattern charts, it may also give separate plots for multiple distances. If no angle is specified, measurements are assumed to be on axis (0-degrees). The particular frequency response characteristics of any microphone are one of the factors that make it distinct from others in its sonic performance.
For the most accurate and least “colored” performance, look for a flat (even) frequency response. While this may initially sound like it would always be a good idea, many highly desirable mics are famous for the particular coloration they impart on a recording precisely because their frequency contour is not flat. The ubiquitous Shure SM58 mic is a great examples of this. They have a distinctive, strong presence peak (a boost from 2.5 to 6.5 kHz), and roll off frequencies above 12 kHz. They also begin to roll off frequencies below 100 Hz, though the influence of their proximity effect often makes them seem strong in the low end. It is this combination of spectral features that makes them such a common vocal mic. (More on this in TCRM #12) Experience, trial and error, and lots of careful listening will help you determine which microphone to choose for a given task.
Use the manufacturer-supplied graphs as a rough guide only. Manufacturers sometimes supply a brief, simplified version of a mics frequency response by stating a frequency range throughout which the unit stays within a certain sensitivity variance. Example: 30-15,000 Hz, ± 3 dB. That means there’s a 6dB swing (far from flat), but beyond that figures such as these are really not too helpful in understanding what a microphone will sound like, or even to accurately gauge its usable range. The example given above might lead to the assumption that the microphone is very weak at 20kHz, when in fact it may only be down –4dB from average: and the response to the extended high end may not matter if you’re close-miking a bass amp, for example.
Another common practice among mic makers is to publish smoothed-out graphs that average out sharp spikes and dips in the frequency response; in some cases, these can include case or grill resonances that affect how a particular mic sounds. To be fair, if these are slightly different for each mic of that type, then a generalized graph couldn’t possibly include them. So as a rule, don’t be guided only by oversimplified specs and graphs drawn on a less-than-detailed scale – let your ears be the ultimate judge.
Another important factor that greatly influences the sonic character of a microphone is how quickly the diaphragm can react to changes in acoustical energy. This ability, known as transient response, is governed by one of the most fundamental of physical principles: inertia.
An object in motion wants to stay in motion. An object at rest wants to stay at rest. The amount of force needed to change these states is directly proportional to the mass of the object.
When a sound wave reaches a microphone’s capsule, it is the mass of the diaphragm that determines how much force will make it move, how far it will move, and how quickly it can react to change. Dynamic microphones tend to react more slowly due to the fact that their diaphragms are heavy and have voice coils attached, adding to the overall inertia. Ribbon designs generally exhibit a faster response as the ribbon element is both lighter and does not have attached wire coils. Newer condenser designs tend to have the fastest response times due to both their mechanics and very thin, light diaphragms. These are sometimes as little as 3 microns thick (.003mm). In contrast, however, some venerable older condensers have 13 micron diaphragms, but their sound is still considered extremely desirable for many applications.
Response time is especially important when recording sounds with strong, aggressive attacks such as percussion or brass. To accurately capture these attacks, a microphone with a fast response must be used. Small diaphragm condensers do this well (see next topic). On the other hand, microphones with slower response times are often used to tame these sharp peaks. For this, dynamics are well suited. While fast transient response may be more accurate, it may not necessarily be more desirable. This is an aesthetic/artistic decision.
Does size really matter?
In this case, it certainly can. Diaphragm diameter, thickness, and materials all influence the transient response and frequency response of a microphone. The overall size and density of the material used determine the mass/inertia, which translates into transient response. For this reason, microphones (esp. condensers) are often referred to as either small- or large- diaphragm. Generally speaking, small diaphragms are those which are less than 3/4-inches in diameter. Large diaphragms are those which are greater than 3/4-inches. Materials and dimensions also influence the natural acoustic resonances of the diaphragm itself. This affects the frequency response of the microphone, especially the off-axis coloration. Smaller diaphragms will have less coloration than larger ones, as a rule – that’s why measurement mics often have tiny diaphragms, ¾-inches or less.
How much is too much?
When an acoustic vibration (sound) causes a mic diaphragm to move, the distance of its movement is determined by the force of the acoustic energy. To oversimplify a bit, the louder the sound, the further the diaphragm moves. It can only move so far until it either distorts (stretches or warps), hits something close to it, or tears. All of these things may not only ruin a recording, but the mic as well.
The circuitry of some microphones will overload, causing the audio to distort, before any of these things occurs. This can give some warning that a mic is nearing its limit, but should not be counted on. For this reason, mics are given a maximum SPL (Sound Pressure Level) rating. As a general rule, dynamic mics can handle higher SPLs than either ribbons or condensers. You should familiarize yourself with the SPL rating of all your microphones if you care about their health.
To become familiar with SPLs, it’s a good idea to purchase a meter. Inexpensive SPL meters are available at Radio Shack. These are useful for calibrating monitoring, checking levels (for the health of your mics and your ears!), and running acoustics tests to diagnose and treat studio acoustics concerns.
As a point of reference, the following sounds and their associated SPLs are all above the human threshold of pain (ca. 125dB SPL) and the maximum SPL handling of many microphones:
- Cymbal crashes - can reach around 130dB SPL at 6 inches.
- Singers - can exceed 140dB SPL when the mic is only an inch away.
- Kick drum – can exceed 150dB SPL with microphone inside.
- Snare drums and trumpets - can exceed 160dB SPL within 2 inches.
On a related note, blowing into a mic to check if it’s working is not a smart move. It can damage the diaphragm by forcing it to move farther than it’s designed to. This is especially true of, but not limited to, ribbon mics.
A microphone’s sensitivity is determined by measuring its output voltage when exposed to a given SPL input. The greater the output, the less amplification a signal will need to reach line level. Because less gain is required less noise is added to the signal. Condenser designs tend to be more sensitive than dynamics or traditional ribbon designs.
Every microphone introduces a certain amount of noise to the signal due to its own mechanical and electrical workings. The amount of noise it generates in this manner is called self-noise or equivalent-noise. This is expressed as the amount of dB (SPL) that would produce an output voltage equivalent to the level of the microphone’s self-noise. 20dBA or less is considered to be very good. (The “A” in dBA means A-weighted, a filter algorithm that stresses mid frequencies to simulate human hearing at lower levels) Anything over 32dB is considered quite noisy, so less is better.
Another way to look at a mic's performance in relation to noise is expressed as a ratio of signal-to-noise. This ratio is expressed as the difference between a microphone’s sensitivity and its self-noise. Here, the higher the number the better, as is the case with a S/N ratio for any piece of gear.
Resistance is futile
A microphone’s resistance to the flow of electrical current (aka. impedance or Z) is expressed in Ohms. It affects the length of cable that can be used without increasing noise and/or hum while losing level and/or higher frequencies. Lower impedances are used to minimize these problems (generally around 150 to 400 Ohms at 1kHz). At the same time, the relationship between the mic resistance and that of the preamp is also important. To ensure the issues outlined above are kept under control, impedances of around 10 times that of the microphone are used.
Recently, people are becoming increasingly aware of the sonic coloration caused by interaction between the mic Z and the preamp Z. This is due to the fact that a number of preamps now come with variable impedances. Lowering the preamp impedance will lower the mic’s output as well as tend to emphasize the natural peaks and resonance’s of the microphone. Increasing it boosts the level, flattens out the low and mid frequency response, and increases the high frequency response. Ribbon microphones are especially sensitive to this impedance relationship; it can be a useful color to add to your palette when recording!
More buttons, dials, and switches
Many microphones come with additional controls. Some allow the user to select from multiple polar patterns. Others include a high-pass filter (a.k.a. low-cut or bass roll-off) to reduce low-end sensitivity. This can be used to counteract proximity effect and to reduce such things as handling noise, mechanical noise, wind noise and plosives (see below).
To control sensitivity to high SPLs, a attenuator pad is sometimes also included on the microphone itself. It can be used to keep the mic and/or preamp from overloading. When it is not needed it should be bypassed to reduce noise. It’s not a good idea to unnecessarily attenuate a signal just to boost it again later.
Microphones for use outdoors or in any sort of air current should be equipped with a windscreen to keep moving air from blowing on the microphone element, causing a rumble in the recording. Some microphones come with foam filters that fit around the capsule, but aftermarket versions are also available.
Pop filters (or plosive filters) work on a similar concept. These are generally made of a fine mesh material (like pantyhose) stretched inside a round hoop. Some fancy ones, like the Stedman Proscreen, are made of metal mesh. The hoop is held in front of a microphone by either a gooseneck or rigid boom arm. Fast moving puffs of air from the singer’s mouth (that come with plosives – sounds like “p” and “b”) are blocked from hitting the microphone element, but sound gets through.
A shockmount is also a good idea as it will help isolate the mic from any ground, floor, or stand vibrations that could generate unwanted noises in the recording. Different models of microphones are more and less sensitive to these types of mechanical vibrations. Some are even internally suspended in a shockmout.
A mic’s a bass roll-off (high pass) filter, if present, may also be used to address these types of issues. Be aware, however, that filtering also permanently removes all low frequencies from the recording, not just those that are unwanted. Be sure you don’t need any of those lows before you axe them.
Well, that does it for now. Next time we’ll discuss practical approaches to microphone choice and placement.
John Shirley is a recording engineer, composer, programmer and producer. He’s also on faculty in the Sound Recording Technology Program at the University of Massachusetts Lowell. Check out his wacky electronic music CD, Sonic Ninjutsu, at http://www.cycling74.com/c74music/009.
Here are recordings of the same electric guitar take from the perspectives of various microphones.
First, compare the Shure SM57 dynamic mic (TCRM8_1) with a Neumann TLM103 large diaphragm condenser (TCRM8_2). They are both placed on-axis directly in front of the speaker about 6-inches from the grill.
Next, two dynamic mics are placed 1-inch from the grill. This time the SM57 (TCRM8_3) is contrasted with an AKG D112 (TCRM8_4).
Now, the same microphone model can be compared in two close locations. The SM57 at 1-inch versus the SM57 at 6-inches. (TCRM8_3 vs TCRM8_1)
Next, the TLM103 is compared at two different locations. First at 6-inches again (TCRM8_2) and then at 4-feet (TCRM8_5)
Finally, the electric is recorded at by two distant mics: a TLM103 cardioid (TCRM8_5) and a Rode NT2 in omni (TCRM8_6).
Now here's a classical guitar recorded using various mics and patterns, but all from the same distance (16-inches). Compare patterns and mic types.
A Lawson LMP47 in cardioid.
A Lawson LMP47 in omni.
A Lawson LMP47 in figure-8.
A Rode NT2 in cardioid.
A Rode NT2 in omni.
A Rode NT2 in cardioid.
A Neumann KM184 cardioid small diaphragm condenser.
An Audio Technica cardioid small diaphragm condenser.
A Shure SM57 dynamic.
An AKG D112 dynamic.