by Marc Röhm

In this article, I will address the benefits of using saturation to control dynamics at the mixing stage of a song. I will illustrate the effects of saturation using studio drum recordings as an example, specifically snare drum recordings, since the snare drum is one of the most important elements that contribute to the feel of a song – and quite often it is also the loudest element in a mix.

Introduction

On the time axis, a sound can be divided in two constituent parts: the transient and the sustain. 

Transients are short bursts of noise at the beginning of sounds originating from chaotic vibration of an abruptly stimulated vibrating system (e.g. a string or an air column). They are sometimes also referred to as “attack” and are caused by the beater hitting the kick drum, the stick hitting the head of the snare drum, or the plectrum hitting the string of a guitar, for example.

Transients are especially prominent on percussive sounds, drum transients being one of the loudest transient species. Here, they occur during the first 10 milliseconds. Transients are essential to articulation and our interpretation of sounds. 

The sustain is the part of the sound that follows the transient. It often contains the harmonic information, like the tone of the guitar string, for example.

Starting The Mix 

When starting to mix a snare drum track, I’d often use a Neve-style EQ. The sound and color of these EQs have been – and still are – widely popular and can be heard on countless records. One of the most classic ones is the EQ in the Neve 1073 preamp module.

I’d use the low shelf of the EQ, set to 220 Hz, in combination with the high pass filter, set to 50 or 80 Hz, to add more body to the snare, while removing unwanted low end rumble. The high shelf and the 7.2 kHz band can be used to increase the high frequencies of the sound in a very pleasing manner.

Mix In Saturation

As you might have noticed, I also increased the microphone input gain on the 1073 preamp. As result, the electrical components of the preamp (transistors, transformers, etc.) become “overloaded” – or saturated: they are presented with a higher amount of input signal (current) than they can handle in a linear fashion and hence, pass the signal in a nonlinear manner. Thus, the outgoing signal is altered compared to the input signal.

The saturation of the electrical components results in two effects:

1. compression and
2. distortion (in this case: harmonic generation)

I. Compression

Upon saturation, the incoming signal is compressed, that is, the louder parts are reduced in volume. Being the loudest part of a percussive signal, this affects mainly the transient. Tthis process is sometimes also referred to as “rounding off” the transients by saturation.

Also read: And Yet Another Article About Compression?

The effect of saturation on the amplitude of transients can be visualized by rendering a signal that has passed the preamp at different gain settings:

Graphical representations of waveforms of a snare drum signal (orange, no saturation) that passed a preamp at increased input gain (+15, +30, +45, +60), resulting in increased saturation of the electrical components (the preamp settings are shown next to the corresponding waveform; plugin: Sonimus Burnley 73). The preamp output volume has been adjusted to equal loudness as measured by a loudness meter (not shown here).

As demonstrated by the rendered waveforms, with increased saturation of the preamp components, the transient at the beginning of the waveform gets reduced in volume (compressed) more and more, until it is practically gone at the strongest saturation. In order to maintain punch and definition of the snare drum (and other percussive elements), the transient should be preserved to a certain degree. Therefore, I would probably go for a setting between 15 and 30 on the preamp drive on this plugin.

The benefits of slightly compressing the transient in this manner are:

•  a more dynamically balanced performance and, more importantly,
•  the peak level of the snare drum track can be reduced while retaining equal loudness and punch, given the appropriate saturation algorithm.

Here, the reduction of the peak level of the snare drum track upon preamp saturation while loudness remains constant is demonstrated at more extreme settings:

Effect of transient compression by preamp saturation on peak level. A) without compression, the snare drum track peaks at -7.5 dBFS (Burnley 73 bypassed); B) compression reduces the peak level by 4.4 dB to -11.9 dBFS (Burnley 73 active), while retaining equal loudness at -24.6 LUFS (measurements were taken in a loop until integrated LUFS readouts were stable; plugin: Youlean Software Youlean Loudness Meter).

A & B

II. Distortion

The second benefit of the nonlinear behavior of the preamp is the generation of harmonics, that is, higher frequencies that were not originally present in the signal.

This is a more musical way of generating high-frequency information than boosting high frequencies not or hardly present in the signal with an EQ, since the generated frequencies are harmonically related to the input signal.

In the case of transistor saturation, harmonic frequencies – “harmonics” or “overtones” – that are multiples of the fundamental frequency, get generated, resulting in a fuller and more complex sound, sometimes also described as being more “gritty”.

The generation of harmonic frequencies can be visualized by sending a sine wave through the preamp and analyzing the resulting waveform with a spectrum analyzer.

Also read: What Is The Sound of Analog?

The generated harmonics appear as additional spikes in the waveform:

A & B

Generation of harmonics by preamp saturation. A) A 200 Hz sine wave representing the fundamental (or 1^st harmonic) of a snare drum without harmonic overtones, preamp in bypass; B) preamp active: additional harmonics can be detected at multiples of the fundamental, e.g. at 400 Hz (2^nd harmonic), 600 Hz (3^rd harmonic), 800 Hz (4^th harmonic) and so forth; (plugins: Sonimus Burnley 73, Izotope Insight 2, Klanghelm VUMT).

Combining both effects

While compression by saturation is an effective way of reducing peaks of individual elements in a mix, harmonic generation aids to retain equal loudness or even to increase the perceived loudness of these elements. Applied over the whole mix (on individual tracks and groups), the difference between the maximum peak level and the average level is reduced, resulting in a more compact – or denser – mix.

This difference between average and maximum peak levels of a mix can also be expressed as the crest factor, measured in dB. Thus, the reduction of peak levels can decrease the crest factor of a mix.

This way, load can be taken off group and mix bus compressors, as well as the final limiter. As a result there would be less compression artifacts, and an overall more transparent and better sounding mix. 

For a more detailed explanation see D. Gnozzi from MixbusTv for mixing with saturation: www.youtube.com/watch?v=E-qpQn8BQ2I; and for crest factor: www.youtube.com/watch?v=lk4D4bMu8uo&t=6s; crest factor explained: www.izotope.com/en/learn/what-is-crest-factor.html)

Final Remarks

Mixing with saturation can be an effective way of reducing peak levels of individual elements or groups in a given mix while retaining equal loudness. This results in a denser mix that has the potential to be turned loud without the creation of undesirable compression artifacts. Here, the effect of preamp saturation, resulting in nonlinear passing of an incoming signal where louder parts of the signal are compressed and harmonic frequencies are generated, was demonstrated by the use of a snare drum recording, the snare drum being one of the most important elements in most musical genres – and due to its strong transients also often one of the loudest elements in a song. Happy mixing!