Lecture 6 – Dynamic processors

The dynamic processors are processors, which are sensitive to the dynamic range. Usually we can say that they don’t care about the frequency range and frequency bands, like the eq (there are exceptions, but we will not talk about them now).

All the dynamic processor is detecting and working on, is the dynamic range.

The 4 major types of dynamic processing are:

1. Compressor;

2. Limiter;

3. Expander:

4. Gate.

The compressor and the limiter are devices, which are reducing the dynamic range. This is how the

compressor

works:

On the top you can see the original file without compression. On the bottom – the same, but processed with a compressor. You can see that the hi-level sound are with a reduced level, and the low-level (like at the end of the file) are with increased one.

So, the compressor is reducing the dynamic range. The sounds, which have big peaks, are reduced, and the weak sounds are increased.

How it works?

First, we set a point, from which the signal is reduced with a certain amount, or ratio.

This point is called

THRESHOLD

As you can see from the picture. there is a place, where the line of our processor is crossing the 45-degree line. That exactly is the threshold point.

All of the signals above the threshold are lowered, and all below are raised.

Thus we have reducing of the dynamic range.

The angle of the compression line to the 45-degree line is called

RATIO.

The 45-degree line is ratio 1:1 . When having this ratio the compressor is not working because all the input signals are equal to the output signals. Because this is valid for all levels, the compressor is not working in this mode.

If the ratio is more than 1:1. then we have a compression, i.e. reducing of the dynamics. (2:1, 3:1, 10:1, 20:1, etc…)

If the compression is infinitive : 1 we have a

limiter.

In this situation nothing will go above certain level.

Here is how looks the limiter:

In practice the compressor is not a straight line. It is more like here: it is passive below the threshold and only if the detector circuit detects a sound above the threshold, the compressor starts to compress the signal. Because doing this the levels are doing down, after the process we make a correction of the level with so.called makeup gain. With this gain the levels raise up again. And finally the low levels are raised and the high – lowered.

Here we can see the graph of the process:

and we can see the process like a waveform:

 1 is the initial level,

2 – the level after compressing, and

3 – the level after compressing and after makeup correction.

SO, till now we know 3 different parameters of the dynamic processor –

1. – threshold

2. – ratio

3.-  make-up gain.

Here we can see how looks the waveform of the limiter:

Keep in mind how equal are all of the peaks in the sound. This is because the ratio is infinitive/0, i.e. nothing can go above a certain level.

Where can we use the limiter? In situations, where the amount of the level should be controlled very strictly – in the radio, in FOH work, in the TV, etc.. In this situations it is a good idea to put limiters on the microphones straight after the preamp to reduce the output level.

Because the compressors are more gentle than the limiters, we often use them in the studio. In the modern recording and postproduction it is important to use compressors, because without them it is not possible to mix the dialog with the music, for example, or to set the desired final levels of our mix.

The next type of processor is the

expander.

With it the dynamic range is increasing, not reducing.

The expander is the opposite of the compressor. After it the dynamic range is bigger, not smaller. We do it by changing our ratio from 2/1, 3/1, 10/1, etc.. to 0,9/1, 0,7/1… So the first digit in the ratio should be less than 1. In this situation we have expansion of the sound,not a compression.

Here we can see how it looks:

We can see that the low levels are smaller, and the big – bigger. Thus the dynamic range is increased.

We use the expander when we need to reduce the amount of the sound in the low levels, for example when we have bleeding in the microphone from another acoustic signal near it. Or some unwanted noise or atmosphere.

Another use of the expander is when we want to correct an overcompressed sound. Although it is a risky business, sometimes it is possible to do it with the opposite parameters of the expander.

Ana the final device in the area of the dynamic processors is the

gate

, or nose gate. It is an extreme form of expander, just like the limiter was an extreme form of the compressor.

The gate works by setting a threshold below which the sound is reduced in an extreme way – it is muted. The gate is absolutely passive to the sounds which are above this threshold, it works only to mute the sound below the threshold level.

We can see how it can be seen in a waveform:

We can see how it cuts the low-level signals completely.

The gate can be used for the same purpose like the expander, but it is more extreme. With it we can mute completely or partially unwanted sounds or noise. We have to be extremely careful with the gate, because if if it is used during a recording, the muted sound can not be restored.

Finally, it is time to explain the final 2 major parameters of the dynamic processors.

They are called

attack time

and

release time.

The timbre of the sound depends much from the transients – the process of changing the levels of different harmonics in the beginning and in the end of the sound. Because of this very often the compressor and the another dynamic processors can ruin this processes, because they smash the levels and because of this the color of the transients is lost after them. This alters the sensation for timbre, and the sound becomes more lifeless and unnatural.

Because of this we can use a little lag in the time response of the dynamic processors, telling them to start and and the process a little after the detection of reaching the threshold level. This helps the transients to pass through the compressor unchanged, because it doesn’t “see” them. This makes the sound more natural and full.

Here we can see the transients of a trumpet and the violin. We can see that in the first 100-150 ms the sound is very flexible, different raising and lowering of harmonics occurs, and it is good idea to keep this part of the sound untouched. We can do that for example with the attack time.

That’s where the attack time and the release time are important. When we select a little lag between the passing of the threshold and the actual time of the start of the compression (attack time) or a lag between the falling of the signal below the threshold and the end of the compression (release time), we have more natural representation of the transients: