CompressionCircuit
A basic model...
http://freecircuitdiagram.com/2009/05/09/dynamic-compressor-self-powered-circuit-design/
It is possible to make a compressor circuit from basic components such as the circuit shown. This compressor is composed of diodes, resistors, capacitors, and a variable resistor. At low input signals the D3 and D4 diodes are not activated and so the voltage at Vout will be affected only by the 470k resistor. Once the input voltage reaches a certain level D3 and D4 will allow current to flow and the signal will be attenuated. The attack time of the compressor (the amount of time it takes for the signal to be attenuated) will be controlled by the time constant of the two capacitors and R2. We would be able to adjust the release time (amount of time it takes the compressor to stop attenuating the signal) somewhat by adjusting the variable resistor. In the audio world, compressors are much more complex than this and are usually active - meaning they have some kind of amplification.
Audio Compressors
http://www.kellyindustries.com/signalprocessors/universal_audio_teletronix_la2a.html
To the left is perhaps one of the most famous compressors ever to be used in the history of audio mixing. The LA-2A compressor has been used on many platinum albums on everything from vocals to drums. A compressor like this one can run anywhere from $2000 to $5000. What exactly does it do? In one sentence audio compressors will turn the volume down on an audio signal once it hits a certain threshhold value. They basically act as very fast reacting volume knobs. With most compressors it is possible to set the threshold where the compressor will start attenuating the signal to anything that we want. Also in most compressors we can set the ratio of compression, the attack time, the release time, and set a makeup gain. The best way to explain is with an example. Say there is a singer that on average sings at a volume of 75 db. However, sometimes the singer will sing much louder causing the vocal track to be much too loud compared to everything else. So we can use a compressor and set the threshhold to 78 db and set the ratio to 3:1. This means that for every 3 db over the threshhold, only 1 db is output. So if the vocalist sang at 84 db the volume on the track would now be 80 db. We could also set the attack and release which would govern how long it would take the attenuation to kick in after the signal went over the threshhold and how long it would keep attenuating the signal after it dropped back below the threshhold. Mindful use of compression in the mixing stage is the biggest difference between a profession and amatuer production.
