audio courses logoCompression 101
by Producer/Engineer Karen Kane

Published at AudioCourses.com October 2003

Using a compressor on a microphone or on a recorded track may seem to be a simple task, however, there’s a lot to learn about the theory of compression. Understanding this theory and why compression is useful will help you in using these devices more effectively.

Compressors are “dynamic range processors” or “variable gain amplifiers” (VGA’s). They affect the dynamic range of sound by varying the “gain” (volume) of that sound. If the dynamic range of music is potentially near 120db (that’s loud!) and the dynamic range of recording devices and live sound systems are only near 75-90db, we need a way to control the dynamic range so that those loud levels do not cause distortion. The dynamic range has to be gain reduced to a level appropriate for recording or a live concert. This “gain reduction” could be accomplished by manually riding the fader up and down but our hands are not fast enough. A compressor automatically controls the volume of a sound, leaving our hands free for other things. Compressors are used on vocals, bass, guitar, drums and almost anything that fluctuates widely. Very simply put, a compressor is an amplifier whose gain decreases as the input increases. It makes big changes into smaller changes.

Most compressors have 4 main controls: input level, output level, threshold and ratio.
Some compressors also have attack time and release time controls.

Input level: is exactly that, the amount of input level. When you set it to “O”, you get the exact level that you are sending from your mic pre-amp or from an already recorded track.

Output level: the amount of output after compression.

Threshold: A set point at which the compressor begins to work. Incoming levels below your set threshold are unprocessed. Incoming levels above your set threshold are compressed according to the compression ratio. The threshold on a compressor is similar to an air conditioner thermostat. When the temperature exceeds a certain threshold, the air conditioner kicks in. When the temperature drops below threshold, the air conditioner shuts off. This is the same idea for the threshold of a compressor.

Ratio: is the ratio of the input signal to the output signal after compression.
Example: 4:1 ratio means if the input is 12db, the output level is only 3db. So level input (12db) divided by ratio (4) is the output level. A ratio of 2:1 would give a 12 db input an output level of 6 db and a ratio of 8:1 would give an output level of 1.5db. So the rule of thumb is, the higher the ratio, the more extreme the compression is going to be. Therefore, ratios from 2:1 up through 6:1 are considered “gentle” and ratios above 6:1 are considered “hard”. “Hard” compressing is called “limiting”. A limiter is merely a compressor with a very high ratio, very fast attack times, fast to medium release times and high thresholds. Limiters are great for live sound systems as a safety device when very high levels are introduced into the system. Any signal above the threshold is “clipped” off. They protect speakers from blowing. 8:1,10:1, 20:1, 100:1 are common limiting ratios.

Attack time: The speed with which the device affects the signal. The time it takes to react to a signal above the threshold.

Release time: The rate at which the device lets the signal decay. The time the compressor takes to return the signal to normal (the way it was before hitting the threshold).

Stereo compressors come in one box with 2 channels and can be used in 2 different ways. You can put 2 individual mics or tracks into it and each channel will have completely independent settings from each other. Obviously, a mono compressor just has one channel. The second way to use a stereo compressor is to put an entire stereo studio mix or the stereo output of a live soundboard into it. This is called using the compressor as a Left/Right Stereo Bus compressor. When you use it as a L/R Stereo Bus compressor, you have to link both sides electronically so the same amount of compression happens on both sides at the same time.

A “Frequency Selective Compressor “(De-esser) is a special compressor that reduces the level of a very narrow band of frequencies. It’s very useful when a singer has a strong, sharp, sibilant “S” to their voice. Typically in the 5-8k range. When 5-8k exceeds the threshold, it reduces strong, sibilant “S’s” without affecting the rest of the word.

There are 2 types of compressors: Tube and solid state. They have a wide range of quality and price ($200-$4000). Upper mid price and expensive compressors can compress a signal heavily and you can hardly tell it’s working. The sound remains much more natural than when an inexpensive compressor is compressing.

Tube compressors have such a great sound that sometimes a signal is put through it just to get the sound of the tubes while hardly compressing it at all.

Be careful not to overuse a compressor. When overused, it creates a very unnatural sound. The trick is to learn to use a compressor in a subtle manner.