#
9.5. USING THE COMPRESSOR/LIMITER
The dynamic range (how loud we can hear to how quiet a sound we can detect) of the human ear is far
greater than the capacity of sound systems to reproduce. Although some of this equipment limitation
is at the upper extreme of the dynamic range (where too loud a signal will produce distortion), much of
the restriction occurs at the low level end, where the signal disappears below the “noise floor” of the
circuitry.
A compressor (or in its most powerful form, a limiter) is the most widely used tool for controlling dynamic
range. In the simplest terms, a compressor is designed to restrict the dynamic range of an audio
program; i.e., to make quiet signals louder, and loud signals quieter. A compressor becomes a limiter
when the compression ratio (the ratio of the input gain change to the output gain change) is so high that
the output level won’t rise above a “brick wall” ceiling regardless of how loud the input gets.
A compressor acts like an “automatic mix engineer” with a hand on the fader, and an inhumanly fast
reaction time. When the input level increases, the engineer drops the fader; when the level decreases,
the fader is raised. If the amount of fader compensation equals the variation in signal level, the output
level of the audio program will sound consistent.
The practical benefits of compression include:
1.
Speaker protection. A compressor will control sudden level peaks and prevent your speakers from
damage.
2.
Perceived increase in loudness. Because peak levels are kept from rising as high as uncompressed
signals, you gain headroom for your audio program and can raise its overall average gain.
Compression is often added to the entire audio mix, both in live sound and recording, to increase
its perceived loudness.
3.
Mix consistency. For expressive instruments or vocals, which may have a large dynamic range,
compression can help maintain consistent mix levels. So a vocal that varies from a whisper to a
scream will not disappear or stand out in the mix, relative to other less dynamic instruments.
Like any signal processing, compression can be misused, and cause undesirable problems in the audio
signal. Some of these problems include:
1.
Noise. If the threshold for compression is set too low, and the output gain is raised substantially
to make up for the gain loss of compression, the resulting output signal can be noisy. This is because
the input signal must be raised significantly to produce the same output level, and the noise floor
of your equipment will be amplified unnecessarily. This problem will be exaggerated if the input
signal level to the compressor is very low (which will already degrade the signal-to-noise ratio).
2.
Breathing. In situations where the compression ratio is high, the threshold is low, and the release
time of the compressor is short, the noise floor will modulate up and down as the audio signal stops
and starts.
3.
Pumping. When the compressor release time is set too long, low level signals that follow a peak
closely will disappear or fade back up as the compression releases. Experiment with the release
time to obtain the ideal setting; this will be a function of the program content.
4.
Over-compression. Applying too much compression to a mix can sometimes result in such evened-
out dynamics that the “life” of the music has been removed or curtailed. Dynamic variation in music
is a major component of its excitement and interest; don’t remove them, just control them. This may
be particularly true for percussive sounds such as drums. While compressing drums can “fatten”
up the sound, setting the attack time to be longer may allow the initial strike of the drum to have more
impact.
Section Nine: Suggestions for Optimal Use of the GRAPHI-Q
