In the past couple of years, interest in acoustic music has reached new highs. Sparked in large part by the phenomenal success of the movie O Brother, Where Art Thou and artists such as Nickel Creek and Alison Krauss, acoustic music is being performed in more and more places nationwide. That has focused attention on the question of acoustic-music sound reinforcement. What is the best way to successfully translate your acoustic instruments into the P.A., so that the audience can experience your music in the most authentic, and sonically pure manner?

In the right live situation, such as a performance by a bluegrass band or other acoustic ensemble without loud drums and electric instruments, using microphones only (rather than a combination of mics and instrument pickups) can often provide the best-sounding solution. You need to use the proper techniques and the right mics, of course; but if you do, the results can be sonically very satisfying.

But why not use pickups? After all, pickup technology has become more sophisticated, and now you can get great sounds and superior gain before feedback, with high-quality transducers and preamps. Well, because despite the sophistication of current pickup technology, microphones still come closer to capturing the true sounds of acoustic instruments. Microphones are overwhelmingly the choice in the recording studio. Try it yourself; record a nice flattop simultaneously to two separate tracks — one with the pickup of your choice, and the other track with a decent condenser mic or even a Shure SM57. The miked track will almost always sound more natural.

Mics have other advantages in addition to their sound. Let's say you play many instruments and have two acoustic guitars and a mandolin that you use for your gigs. Outfitting all three with pickup systems, preamps, equalizers, and direct boxes could be a very pricey proposition. You could spend less by buying a decent-quality condenser mic that could be used to mic the lot. That would also mean that you wouldn't take up as many inputs on your band's mixing board, which could allow you to use a less expensive, physically smaller mixer.

Imagine this: for a five-piece acoustic band in which everyone sings, you could effectively mic the entire band with ten mics, which can be set neatly in the front line providing a clean, wire-free stage area in which changing instruments is as easy as, well, changing instruments. This approach requires no switching, muting, or repatching, and the untethered musicians are free to roam the stage and venue. Performers can learn to “work” the mic (move back and forth from it) to control volume, dynamics, and tone.

Mics do have limits. Strong winds can generate a roar in them (even if they're outfitted with foam wind screens) that can require the filtering out of frequencies as high as 200 Hz (about the frequency of a guitar's open-G string) and can literally blow an instrument's sound off course (and thereby reduce the level coming into the mic) before it reaches a mic.

As mentioned, loud drums can present a problem for miked acoustic instruments, raising the overall stage volume and making it necessary for the monitors to be louder. That, in turn, can create feedback problems with all those open mics. That said, drums can be played more quietly with brushes or blasticks (depending on the musical style), and a plexiglass drum shield can be used to direct drum sounds up and away from the front-line mics.

Even with the problems that volume can bring, I have balanced a quiet oud and saz with multiple hand drums and gotten the “chunk-chunk” rhythm of an archtop guitar to cut through a 17-piece big band with a mic at the lower “F” hole. It can be done, if you know how to do it. But before getting into specific applications, let's review some microphone basics.

Most stage mics are either dynamic or condenser. Dynamic mics offer rugged, reliable construction and a high performance-to-cost ratio. In fact, the use of lightweight neodymium magnets has allowed some dynamic mics to offer performance close to that of condensers.

Condenser mics offer excellent transient response and detail, extended high end, high output, and greater “reach” than dynamic mics. Once considered too fragile for gigging, a multitude of road-rugged condensers have been brought to market in recent years. Unlike dynamic mics, condenser mics require a power supply, which can take the form of an onboard battery, inline power supply or, most commonly in sound reinforcement applications, phantom power supplied by the mixing console via the mic cable.

Mics are designed with one of several possible pickup patterns (aka polar patterns) which determine how sensitive they are to sound coming from different directions. Just as higher frequencies beamed from an instrument or loudspeaker tend to be more directional than lows, mics tend more towards directionality at higher frequencies. Generally, the higher quality the mic, the more uniform the pattern at all frequencies.

The most common polar patterns for sound reinforcement are cardioid, hypercardioid, and supercardioid. A cardioid mic is most sensitive on-axis (the direction in which the mic is facing), about 6 dB less sensitive 90 degrees off-axis (directly to the side), and about 20 dB less sensitive 180 degrees off-axis (opposite to where the mic is pointed).

Supercardioid mics are less sensitive to the sides than are cardioid mics, but have a small lobe of sensitivity at 180 degrees off-axis. They are most insensitive at 120 degrees, as are hypercardioid mics. Hypercardioids are even less sensitive at 90 degrees than supercardioids, but more sensitive at 180 degrees (see Fig. 1).

I favor the use of supercardioid and hypercardioid condenser mics for individually miking stringed instruments. The greater directionality of these patterns can significantly improve isolation and gain before feedback, and make it possible to use a mic a few inches farther back from an instrument, further reducing the possibility of feedback caused by reflections off the instrument into the mic. It also allows the player more freedom of movement and lessens concern about accidentally bumping the mic.

The working distance (the distance between a mic and a sound source) and the point on an instrument at which a mic is aimed have a major effect on the instrument's tone as projected by the loudspeakers. Acoustic instruments produce complex sounds, with different aural components radiated by different parts of the instrument. (Aiming a microphone at the 12th fret of a guitar yields a very different tone than when the mic is near the soundhole.) Pulling the mic back allows it to sample a greater part of an instrument's area and therefore capture more components of the overall sound. Distance also reduces the proximity effect, which increases bass response as distance to the source decreases, and which can cause guitars to sound boomy.

A pro-quality hypercardioid condenser mic can cost two or more times the price of a professional cardioid dynamic, with neodymium magnet dynamics somewhere in between. Because you may not be able to afford your first choice for every mic needed, use the most feedback-resistant mics for the quietest or most difficult sources.

Although it's preferable to use hypercardioid or supercardioid mics when miking acoustic instruments, there are some instruments, such as fiddle, banjo, and mandolin, that project so well that you can get good results from a cardioid dynamic such as the widely used Shure SM57. (So popular is the 57 that some performers insist on them no matter what other mic is offered).

Violins may be miked from a distance (anywhere from about six inches to two feet), but fiddles blend better with a close-miked bluegrass band when they're close-miked. F-model mandolins produce the round sound preferred by many bluegrass players when miked at the F-hole.

Flat-top guitars, on the other hand, are not nearly as easy to mic, especially when softly fingerpicked. Loudest at the soundhole, the sound there tends to be boomy, unbalanced, and undefined. The AKG C 1000 (a condenser that can be configured with a hypercardioid pattern) gives excellent results when aimed where the fingerboard overlaps the body.

I've also had good results using the C 1000 on resonator guitars. I've gotten a balanced tone when placing the mic between the cone and sound-holes, but greater volume at the cone. For double bass and cello I've had good luck with AKG's large-diaphragm C 3000 (not the cardioid-only C 3000B), which works wonders when aimed at the treble-side F-hole. The C 3000B also works well at the end of the fingerboard for slapped [upright] bass, shines on guitar, and is superb for choral pickup. Bass and cello are also well served by large diaphragm dynamics such as Sennheiser's classic MD 421.

Hammered dulcimer is pretty loud and should be easy to mic, but flying hammers may force the mics farther from the instrument than you might like — nobody wants their nice condenser mics to get whacked. These instruments have more shimmer when miked from above and a warmer sound underneath. Harps will work well with a hypercardioid condenser, aimed one-third of the way up the soundboard, positioned just far enough away so that the harp and mic don't collide when the instrument stands on its base. Sitar, sarod, oud, and saz are quiet instruments, and want to be miked as close as the player will allow. I typically use a C 1000 or C 3000 on those instruments.

The Neumann KM185 has proven a superb choice for the plucked string player willing to invest and care for a premium personal mic, as has the large-diaphragm, multi-pattern AKG C 414 B-ULS (set to hypercardioid). I've yet to test the AKG C 480B/CK63 and SE 300B/CK93, but their specifications look promising. On a beer budget, I've gotten good results with the ribbon dynamic Beyer M-500 and neodymium magnet Shure Beta 56. (Please note that the mic recommendations I make in this article are based on my own experience as a sound engineer. There are doubtless other mics from other manufacturers that will perform well in the applications discussed here.)

Budget permitting, an alternative approach is to use matching mics on the entire band. Using a single mic model introduces less frequency response anomalies to the system than multiple mic types, making system equalization — which will be discussed later — less difficult. Monitor placement is also simplified with only one polar pattern to consider.

Choosing the right mic is especially important in this type of scenario. In this application, a colleague of mine favors the AKG C 535 EB condenser. Even though it's a cardioid and lacks the aforementioned benefits of mics with narrower polar patterns, it's a lovely vocal mic with a detailed, airy sound, and it can also be used effectively on instruments. My suggestion for a less-expensive mic is Shure's supercardioid dynamic Beta 57A.

But just choosing the right mics and placing them correctly is not enough to ensure a successful outcome. You must also make sure that your speakers and monitors are correctly positioned, and that your system is properly equalized.

Every time you add another live mic into a stage setup, you're increasing the possibility of feedback. Therefore, it's critical to set up and tune your system correctly.

Feedback is generated when sound from one or more loudspeakers is picked up by a mic and re-amplified. Mics and speakers ideally should be oriented so that the most insensitive part of the mic, known as the null point, faces the monitor speaker. A cardioid mic's null point is 180 degrees off axis, so the preferred monitor placement would be directly in front of a performer. For super- or hypercardioid mics, however, the ideal monitor position is at 120 degrees off axis (see Fig. 2).

If the performer is at 12 o'clock, the monitor wants to sit at 4 o'clock and/or 8 o'clock. When a performer uses both a dynamic cardioid vocal mic and a hypercardioid condenser instrument mic, monitor placement is determined by which mic needs the most gain. A guitar's mic will need to be cranked, therefore the monitor should be positioned at 120 degrees or the null point of that mic. A loud instrument such as an accordion, on the other hand, rarely needs much level in the monitor. Therefore, you can position the monitor directly in front of the vocalist for optimum gain before feedback.

If only two monitors are being used for an entire band, it's preferable to position the wedges slightly to the front and to either side of the performers, toed-in enough that the musician in the center of the front line is able to hear the highs. Positioning the speakers while listening carefully should allow you to constructively combine the monitor's output at the center of the front line so that they seem as loud in the center as they do to either side (see Fig. 3).

It's a 3-D world, so we must also consider the vertical axis. Aiming the mic approximately straight back and parallel to the stage floor should keep the monitors close to the mic's null point in all three axes (vertical, horizontal, and front-to-back), and helps avoid pointing the sound-sensitive backs of your super- and hypercardioid mics at the mains, as well (see Fig. 4). Because people and, especially, the big, flat top of a flat-top guitar reflect sound, having the monitors at an oblique angle helps sound waves to bounce away obliquely, rather than straight back into the microphone.

It's also very important to position the main speakers correctly. First and foremost, it's impossible to over-emphasize the importance of elevating them. At minimum, the horn of a 2-way speaker enclosure should be raised to seven feet. This allows your speakers to cover more space with less output by projecting the music over the heads of the audience, thus minimizing sound absorption. Because directionality of sound increases with frequency (the higher the frequency, the more directional the sound), lower notes can “wrap around” the speakers and get into your microphones, causing feedback. The lower the output of the mains, the less wraparound, and consequently the less feedback.

Speakers of up to 100 pounds can be safely mounted on an appropriate lightweight aluminum tripod stand, the cost of which is insignificant compared to the improvement in performance it provides. Purchasing speaker stands will buy you infinitely more sonic improvement than applying the same amount of money to a more expensive speaker.

The worst place for your main speakers is behind the performers, where they project directly into your mics, thus greatly increasing the risk of feedback. Sometimes you will be asked to place them against a back wall, but off to the sides, which is better but still not very good. The best locations will be to each side and slightly in front of the performers. Often, you won't be able to place them in front because of sight-line or traffic issues, and you'll have to make due with placing them even with the front line of the band.

Avoid aiming your speakers directly at a wall opposite the stage. If the wall is close, reflections can be loud enough on stage to cause feedback. At certain distances, the reflections can be perceived onstage as a “slapback” echo and thus interfere with the performers' timing. “Toeing in” the speakers at an oblique angle can diminish any echo effect on the stage.

Because loudness diminishes with distance, the likelihood of feedback diminishes when you increase the distance between the speakers and mics. By adding a little distance (roughly 2 to 4 feet, depending on the venue) between the speakers and the side of the stage, you can improve gain before feedback. However, placing your speakers too far to the sides can leave a “hole” in your audience coverage down front and center. Moving the speakers still farther out can destroy the perception that the music is coming from the stage — it will sound as if it is coming from the speakers instead.

When a sound system interacts with an environment, pretty much everything — room dimensions, furnishings, occupants, loudspeakers, microphones, even the weather — affects the relative loudness of low notes, high notes, and overtones. Any tone prone to prominence will likely be prone to feedback as well. Therefore, proper equalization is key when miking acoustic instruments. Graphic equalizers use slide potentiometers to allow the operator to raise or lower the volume of ⅓-octave, ⅔-octave, or one-octave slices of the frequency spectrum.

Octave-band EQs (usually 7- or 10-band) are often built into powered P.A. heads and mixers, while ⅓- and ⅔-octave EQs are typically outboard rackmount units. The more bands an equalizer has, the narrower those bands are, and thus the greater the ability to eliminate feedback while affecting the overall sound less. I keep a chart taped to my mixers which shows the frequencies of the notes of an extended piano keyboard and relates them to the industry-standard center frequencies of the graphic EQ bands. Below 7.9 kHz you can use a piano to find the nearest note to the feedback frequency, check your chart, and pull down the correct band. With time and practice you can learn to identify which band is ringing, and figure out which monitor or mains mix is causing a problem during a show.

It's preferable to attenuate or eliminate troublesome frequencies before the show starts, of course. This is where the process of ringing out a system — intentionally causing problem frequencies to feed back and then attenuating them — comes into play. It should be done with consideration of the likeliest causes of feedback. Quieter sources require more gain and thus pose a more substantial feedback hazard.

System tuning for a mic-intensive string band is best accomplished during sound-check with all performers in their playing positions, simulating as closely as possible the actual performance conditions. Ringing out the mics on a bare stage will yield a totally different result than with the band onstage.

The performers should be quiet during the process so that the operator can hear the slightest onset of feedback. When the performers are not available, I simulate their presence using other available persons, music stands, cases on chairs, or whatever else is available. To protect their hearing during the ringing-out process, I ask my subjects to hold their ears, or I have them wear headphone-style hearing protectors (available at gun shops).

I also usually announce to all within earshot that the next few minutes may be unpleasant, and that they may wish to be elsewhere. Preparing for festival-style presentations, I tune the system after placing live mics and reflectors all about the stage.

I start by using the graphic EQ to drop out all frequencies below 40 Hz, or below the frequency range of the speaker in question, in order to avoid wasting amplifier power and possibly damaging a speaker trying to amplify irrelevant lows. With mics on, and set to their approximate levels relative to one another, I bring up the overall volume until feedback starts, then lower it just below the threshold.

I slowly raise each band, being careful not to let feedback howl uncontrolled. (A compressor/limiter placed after the EQ and before the power amps, and set so as to prevent amplifier clipping, can be useful in protecting your speakers from uncontrolled feedback.) If I can raise a fader to the maximum amount of boost available without causing feedback, I bring it back to unity gain (no boost, no cut) and move on to the next. If feedback occurs, I cut the band in inverse proportion to the boost needed to cause it.

Let's assume that your equalizer can cut or boost 12dB. If feedback occurs with a 3 dB boost, cut that band by 12 dB. If it takes a 6 dB boost of gain to ring, cut 9 dB; cut 6 dB if a 9 dB boost causes feedback, and 3 dB if it takes a 12 dB boost for feedback to start. Temper these guidelines by considering how long it takes to generate feedback in a particular frequency band. If the howl rushes in quickly, attenuate that band more. If it creeps up slowly, attenuate less.

When adjacent frequencies cause the same note to feed back, determine which band causes the ringing with the least boost applied. Attenuate that band and restore the other(s) to flat. Occasionally, the ringing will be right between two bands. In that case, cutting both bands by a moderate amount will do the trick.

Ring out the mains and each monitor mix separately. After EQing a mix, you should be able to raise its level several dBs and have it sound good still. If the sound is lacking, or if you have cut more than 1/3 to 1/2 the available bands, you have probably over-equalized, trying to get more gain than is reasonable. In that case, return everything to flat and start again.

Listen also for boominess and “overhang.” Sometimes a particular frequency is not overly loud, but has a longer reverb time than others, and lowering the appropriate band can tighten the sound and improve focus. When the process is complete for each mix, listen to the mixes together to see how they combine on the stage. The various mixes should support and strengthen one another. If the combined sound is hollow sounding or thin you may need to re-equalize or reverse the polarity of one or more speakers.

Polarity is most easily manipulated if the speaker outputs on your power amp(s) have double-banana connectors — simply reverse the pins of the male connector. In the absence of double-banana connectors, ensure that all your speakers are of uniform polarity before you take them to a performance.

If you'd rather not do a manual ring-out, there are outboard devices that can help tune your system. A Real Time Analyzer (RTA) listens with a measurement microphone and shows the relative energy in each EQ band. You then play pink noise over the PA, and the RTA's display shows you which bands to attenuate, and by how much, to achieve flat (and, theoretically, feedback-resistant) response. During a performance, loud feedback will show clearly on an RTA, but a subtle ringing will still need to be identified by ear. There are digital processors from companies such as Sabine, dbx, and others that combine the RTA with a graphic EQ to tune a system automatically. Flat response, however, particularly as determined by a machine, does not necessarily equate to good sound. Your (hopefully trained) ear should be the final arbiter.

There are also automatic feedback filters from various manufacturers that apply filters as narrow as 1/60 octave to frequencies that their silicon brains believe (sometimes incorrectly) are feeding back. Such devices vary widely in price and performance, and may execute other functions as well. Some work on entire mixes while others are dedicated to a single mic.

For the acoustic ensemble, the mics-only approach to sound reinforcement is by no means the path of least resistance, but it is a viable option, given the right situation. If you learn the necessary techniques, you can treat the audience to a performance that sounds rich, detailed, and right.