Mono-Compatibility, EQ & Compression

Stereo-Miked Acoustic Guitar Tracks and Mono-Compatibility

When stereo tracks are collapsed to mono, the result can sound dramatically different from the original tracks. This is not only true of the listener's perception of the width of the stereo image and the discrete placement of different elements in a mix: Conversion to mono can also significantly change the timbre of individual instruments, especially if certain stereo miking techniques were used to record the original. (In rare cases, the instrument could even disappear from a mix, should the left and right signals be far enough out-of-phase to cancel each other out when combined to mono.)

True, in the last couple of decades -- with AM radio's diminishing role as a music medium -- mono-compatibility has become less of an issue. And in fact, some engineers prefer not to compromise their stereo tracks or limit their recording techniques to cater to the lowest common denominator. Still, many engineers choose to play it safe -- ensuring that their miking (or other processing) techniques won't sound terrible if played back in mono. However you feel about this issue, you'll be able to make more informed choices on how to record if you know what the sonic repercussions will be for mono playback. So what exactly happens to stereo acoustic guitar tracks when they are collapsed to mono? The answer depends on what miking technique you use to record.

Spaced pair techniques generally pick up a high degree of uncorrelated signal for each track. That is, many frequency components on the left-panned track are more or less out-of-phase with corresponding frequencies on the right-panned track. (This still applies even if you use the 3-to-1 rule for mic placement.)

This is because each of the sound waves emanating from the guitar will arrive at each mic at a different time. Due to that time difference the mic will pick up the wave at a different phase of its cycle. When the stereo signal is collapsed to mono, the constructive and destructive interference of these out-of-phase components combine to emphasize and de-emphasize their corresponding frequencies. The resulting timbre can be markedly different from your carefully crafted stereo tracks and can play havoc with your mix. You might be able to compensate with EQ, but this might exacerbate the problem.

Tracks recorded with an X-Y technique are far less prone to phase problems. Since the capsules are placed so close together, the sound reaches both mics at roughly the same time. As a result, tracks recorded in X-Y stereo are much more mono-compatible than those recorded with a spaced pair.

You can hear these sorts of phase relationships by listening to the audio samples linked below. The samples are grouped in pairs playing the same track recorded in stereo and then collapsed to mono. The first examples ("Spaced Pair" and "12th Fret + Right Ear") were produced using spaced pair techniques. You'll notice that the spectral balance -- or timbre -- of these stereo tracks does not survive the conversion to mono very well. The mono versions have exaggerated peaks and dips in the frequency response. Compare this with the X-Y examples: Note how much more consistent the timbre is between the stereo and mono versions. Unlike the spaced pair recordings, the mono and stereo X-Y samples sound quite similar.

A Word on EQ and Compression

Aside from the occasional use of a microphone's passive high-pass (bass roll-off) filter, I prefer not to add equalization to the signal I'm recording. You can never tell in advance exactly how yet-to-be-recorded tracks will interact with the guitar you're recording, so any processing you add during recording is just a guessing game. Since you'll probably need to make subsequent adjustments in timbre (and possibly dynamics) at mixdown, you should avoid processing the signal twice. Your tracks will sound more pristine if you hold off adding processing until you have a complete picture of how the tracks will fit together. Instead, if you're not getting the sound you want during soundcheck, move the mics around until the timbre sounds right.

If you want to experiment with EQ as you're tracking, you can record the music dry and add EQ on the monitor returns. This way, you can hear the results of the EQ without committing to it.

Though many engineers will compress an acoustic guitar during tracking, I usually don't like to do so. As with other broadband, percussive instruments, guitar can easily cause a compressor to pump (cause audible changes in level) if it's not set up exactly right. Once these amplitude modulation artifacts are on tape, they are all but impossible to remove. For this reason, I compress acoustic guitar tracks at mixdown -- when I have multiple opportunities to get it right.