Ken Dibble takes a look at the pros and cons of miking this most expressive of instruments.

All too often the approach to miking this particular instrument is dictated by the microphone types available to hand, by physical constraints imposed by the particular situation, or by ingrained prejudices on the part of the musician, and is always at the mercy of the experience or otherwise of the recording engineer or producer concerned. This is a great pity, because the guitar is among the most expressive of instruments, and whilst the fundamental note or crunch of a chord is easy enough to capture, the finer nuances due to individual playing technique or to individual instrument characteristics are not, and are all too often lost among the rich and heavily emphasised overtones generated by body resonances, thus losing the individual identity of both instrument and player in the general wash of sound.

So let's take a look at some of the pitfalls and see what can be done to overcome the difficulties.

Understanding The Instrument

Before an attempt can be made at capturing the sound produced by any musical instrument, it's necessary to understand the basic characteristics of the instrument itself. In the case of a guitar, low frequency radiation is essentially omnidirectional, with marked peaks at selected frequencies in the vicinity of the sound hole, whilst at mid and high frequencies, most of the energy is radiated by, or is reflected off the sound board, the forward radiation angle decreasing with rising frequency. Always remember however, that it's the attack transients which will provide the fine detail. This information is a function of the way in which a particular musician actually strikes and damps the strings and lies mainly in the upper frequency bands. It therefore follows that as the player's hand will invariably be interposed between the listener and the point of contact between fingers or plectrum and string, what is actually heard must be due to reflections.

As an experiment, place a microphone in front of the sound hole of any guitar, strike a good resonant chord over all six strings, perhaps an E or C at the first position, and move the microphone between the sound hole and bridge whilst the chord is sounding. Over the sound hole the sound will be rich and mellow whilst near the bridge it is thin and mechanical. This is due partly to the low frequency emphasis of the guitar body and partly to a decrease in amplitude of the vibrational modes of the strings at the bridge position.

Basic Approach

In a studio situation, the engineer is freed from the constraints both of howlround and of the musician's requirement that he be free to move about in order to add physical expression to the performance, invariably the case in a live concert venue. Also, the studio engineer does not have to contend with masking due to high energy percussion and amplified instruments in close proximity to the acoustic source of interest. These aspects usually dictate the use of contact mics, or other forms of transducer device mechanically coupled to the sound board or bridge of the instrument for live performance applications. The recently introduced Accusound transducer, for example (which does not require an external pre-amp or power supply), can be fitted directly to a guitar, but good though some of these pick-ups are, there is no doubt that the tonal qualities of the instrument are modified and that expressive subtleties are not captured by such devices. The result is that traditional microphone techniques are the only recourse for discriminating engineers.

It could be argued that in a live performance situation there's more room for compromise because anything missing from the accuracy of the sound is more than compensated for by the atmosphere of the performance itself, and that the trade-off is therefore acceptable. But in the case of a recording, there's only the sound to create the desired stimulation in the listener. There's no action or sense of participation, and therefore, a vivacious and elucid sound quality is paramount. So how do we obtain the necessary accuracy?

"...a good condenser omni, placed about 20 to 30cm out from the sound board, just above the sound hole, as shown in Figure 3 will usually produce an excellent overall balance between radiated and reflected sound and capture the natural tonal characteristics of the instrument."

Tools Of The Trade

The majority of smaller studios rely on general purpose dynamic cardioids as the backbone of their microphone stock with a few condensers thrown in as finance allows. Let's take a look at a typical response curve of such a microphone as given in Figure 1.

Firstly, note that the frequency response is not flat. The lift in the upper mid/treble region is known as the 'presence lift' and is intentionally introduced to provide a degree of brightness to the sound quality. Note also that the low frequency response changes at different working distances, ranging from a marked bass roll-off effect when working in the far field of the microphone, to a pronounced bass emphasis when working close up. This behaviour is known as the 'proximity effect' and applies to practically all cardioid type microphones, dynamic or condenser, unless they are of a type specifically designed to minimise this effect.

When such a microphone is placed close to the strings and over the sound hole of an acoustic guitar (which must be the favourite position amongst musicians and engineers alike) a coupling effect takes place between the body resonance of the guitar and this proximity effect characteristic. This produces a large response peak somewhere below 100Hz, immediately introducing severe tonal colouration due to the resulting low frequency emphasis.

That leaves three choices. Either move the microphone further away from the instrument where it's not within its proximity effect range, move it away from the sound hole to minimise the coupling effect, or use a different type of microphone altogether.

"...no two instruments and no two players will sound alike, so some experimentation has got to be the name of the game."

The first option is fine if the background noise in the studio and desk noise is sufficiently low so as to allow the additional gain without compromising the signal-to-noise ratio of the recording, but even so, it's unlikely that sufficient of the reflected attack characteristics will be retained beyond about 20 centimeters from the sound board.

The second option can often provide a very satisfactory compromise and a suggested alternative position, just above the sound hole at the top of the neck, with the microphone angled so as to reduce standing wave reflections from the neck or sound board. This is well away from any playing position where it might hinder the performer, [and] is shown in Figure 2.

The third option is to change to another type of microphone, the choice being between one of the special cardioid types designed not to exhibit the proximity effect characteristics, or a microphone with an omnidirectional polar response. These are inherently immune to proximity effect considerations by virtue of the fact that it does not work on the pressure gradient principle. In fact, a good condenser omni, placed about 20 to 30cm out from the sound board, just above the sound hole, as shown in Figure 3 will usually produce an excellent overall balance between radiated and reflected sound and capture the natural tonal characteristics of the instrument. If an omni is selected however, it should be realised that it will not only be free of the proximity effect characteristics, it may well also not offer the presence lift effect of the cardioid either, and a dB or so of desk EQ at around the 2.8kHz mark may be required to brighten up some of the reflected transient information, to compensate for the increased working distance.

If a little more warmth is desired, a similarly placed condenser cardioid may be just far enough from the sound hole to provide sufficient low frequency enhancement, without undue colouration. And of course, the desk low frequency EQ can always be used to fine tune for the optimum balance between fullness of tone and colouration.

I have come down in favour of condensers in both cases because of their marked openess and articulation over the mid and high frequency bands when compared to the more conventional dynamics. To prove the point, you need only try an A-B comparison between similarly specified and similarly priced dynamic and condenser microphones for yourself to appreciate the difference, and I'll bet you won't go back to dynamics for guitar work again!

Play it by Ear

Always remember that no two instruments and no two players will sound alike, so some experimentation has got to be the name of the game. But at least now you have some idea as to how the microphone you are using is likely to be behaving, and what the instrument itself is likely to be doing. So you shouldn't now be stabbing in the dark hoping to come up with something that is at best acceptable, and at worst, an insult to the musician and instrument alike!