In audio terms, a stereo delay is a device which allows a time-delay to be inserted into the left and right channels of an audio signal.

If the delay time is equal on both sides, then the result will simply be that you will hear the audio later than is it generated.

Where the delay is different on each side, things start to get interesting, as we will see.

A very short delay (0 to 600us) will effectively pan the sound to one side or the other. This is due to the geometry of the ears. Being about 20cm apart, if the sound arrives late in one ear by the same amount of time it would take to travel 20cm, then we hear it panned far to one side; even if the volume in each ear is the same. Between 0 and 600us, we hear it partly panned to the side.

An explanation of this phenomenon: Imagine, if you will, a triangle. Your head forms the base, and a sound source is at the apex. If you move the apex in a circle around the base, the lengths of the two lines alter relatively to each other. The minimum difference in lengths is 0, and the maximum is the length of the base.

Once we go above 600us of delay, we perceive very little difference until we reach 10ms. At this point, the sound begins to come back to the centre of the image, but now has an added dimension. This is known as a fat or phat sound.

Phatness continues until the delay reaches about 50ms, at which point we begin to hear the sound as two unique entities; and can conciously detect the delay.

With short delays, which are different on each side, and continuously moving, we get a phaser or chorus effect.