Recording - March 2012 - 42

room in each song, a one-of-a-kind personality was lent to each track. This is because no two rooms sound exactly the same and hence studios are frequently chosen for their specific acoustics. Applying this same ideology to lowerbudget recordings can be satisfying as well, regardless of what sort of acoustic space one might have to work in. All rooms are unique, and sometimes even the modest bedroom studio can muster an original sound with minimal treatment. In today’s era of cookie-cutter plug-ins, virtually everyone has access to the same reverb algorithms, and a honest-to-goodness acoustic space can make things stand out. Moving beyond the near field and capturing the room can yield natural and creative drum, horn, and guitar amp sounds.

not without its woes. Although each mic might sound great when heard alone, when combined, differences in sound arrival times create phase cancellation and frequency response suffers. Why does this occur? Let me explain... In last month’s Fundamentals, I discussed how sound propagates in waves of oscillating air pressure. When graphed, a simple single-frequency sound takes the shape of a sine wave, with positive pressure compressions and negative pressure rarefactions clearly visible. Real world sounds are rarely simple sine waves though, as most sounds contain many frequencies and graph as the complex waveforms we typically see in digital audio workstations. (This concept is discussed in detail in the companion video to last month’s article found at Recordingmag.com.) It is important to understand that although graphed waveforms appear static, they depict sound waves propagating outwards from a source. In the course of this movement, waves span distance, and wavelength is the distance it takes for a frequency to complete its entire compression-rarefaction cycle. Since the speed of sound propagation is constant, frequency determines wavelength. The equation is simple: Wavelength = Speed of Sound divided by Frequency.

highs. Although sound travels from a source at a constant speed, every frequency in the sound (the fundamental and each individual the overtone) has its own wavelength. This means at a specific distance from a source, every frequency in a sound is at a different stage in waveform development. How does this mess with sound quality? Put as simply as possible, when near and far microphones are electronically mixed, their signals are out-of-time, and hence the phase relationship of their waveforms is out of sync. Complicating things, since real world sounds are actually a mixture of many frequencies, the phase difference between the close and distance mics varies with frequency. Between close and far mics some frequencies will be “in-phase” and added together, increasing level. Other frequencies will be “out-of-phase” and subtracted. Thus, when close and far mics are mixed together, frequency response becomes uneven, with regular boosts and cuts appearing across the harmonic spectrum. When graphed on paper, this response looks like teeth on a comb, and it is referred to as comb filtering. Two variables affect the severity of comb filtering. First is the distance between microphones, as the greater the separation, the lower the frequency of the

In the top picture, the blue and red waves are identical, and they sum to make a wave with the same shape but greater amplitude.

In the center picture, waves of identical shape but opposite polarity (180 degrees out of phase) cancel out.

In the bottom picture, two identical waves that are time-delayed (slightly out of phase) will cancel some harmonics and reinforce others, causing a change in timbre.

Phase, wavelength, and comb filtering To achieve the best of both worlds, many engineers use both close and room mics when recording drums and other instruments. This technique allows the blending of tight, punchy near field sounds with the realism and ambience of distance mics. While this may seem like the perfect solution, it is
42
RECORDING MARCH 2012

Using an approximate 1130 feet per second as the speed of sound, a few quick calculations reveal that 100 Hz, a low frequency, has a wavelength of 11.3 feet, while 10,000 Hz, a high frequency, has a wavelength of 0.113 feet (about 1.4 inches). These numbers show the obvious—low frequencies have longer wavelengths than

lowest response dip. This often translates into a more obvious reduction in low-frequency level as mic separation increases. Note that at more than twenty feet, phase cancellation becomes less of an issue and farther mics instead seem to capture discrete slap-like echoes. The effects of comb filtering and echoing usually makes it

http://www.Recordingmag.com

Recording - March 2012

Table of Contents for the Digital Edition of Recording - March 2012