13
Faders
Even before the invention of the multitrack, various microphone sources were balanced using faders and summed before being recorded. The summing devices used in early recording studios had faders all right, but these were rather bulky rotary knobs, often 3.5” in diameter. When trying to balance more than two microphones, hands had to jump between knobs. One engineer to foresee the advantage of employing linear faders was the legendary Tom Dowd. By building a control section with small linear faders, he could control each fader using a finger, and so balance more than two microphones at a time. Technically speaking, a fader is a device that can fade sound, whether it be rotary or linear. Today, the term fader is widely associated with linear controls, while pots denote rotary control. This book follows this convention.
Faders are the most straightforward tools in the mixing arsenal. They are the main tools for coarse level adjustments, while both equalizers and compressors are used for more sophisticated or fine level control. Faders are the first thing we approach when starting to mix, and sometimes part of the final fine level adjustments. It might seem unnecessary to say anything about faders—you slide them up, the level goes up; slide them down, the level drops. But even faders involve some science worth knowing. They are not as straightforward as one may think.
Types
Sliding potentiometer
The simplest analog fader is based on a sliding potentiometer. The amplitude of analog signals is represented by voltage, and resistance is used to drop it. Inside a fader, there is a resistive track on which a conductive wiper slides as the knob is moved. Different positions along the track yield different resistance and thus different degrees of level attenuation. One shortfall of the sliding potentiometer is that it cannot boost the level of audio signals passing through it (however, the potentiometer can be made to behave as if boosting by placing a fixed-gain amplifier after it). While the actual circuitry involved is more complex, we can consider this type of fader as one into which an audio signal enters and leaves, as shown schematically in Figure 13.1.
VCA fader
A VCA fader is a marriage between a voltage-controlled amplifier (VCA) and a fader. A VCA is an active amplifier through which audio passes. The amount of boost or attenuation applied on the signal is determined by incoming DC voltage. The fader, through which no audio flows, only controls the DC voltage sent to the amplifier. Figure 13.2 demonstrates this.
Figure 13.1 A schematic illustration of a sliding potentiometer.
Figure 13.2 A VCA fader. The fader only controls the DC voltage sent to a VCA. The VCA is the component that boosts or attenuates the audio signal as it passes through it.
One advantage of the VCA concept is that many DC sources can be summed before feeding the VCA. A channel strip on an analog console can be designed so many level-related functions are achieved using a single VCA. This shortens the signal path and reduces the number of components in it, resulting in a better noise performance. SSL consoles, for example, are designed around this concept—DC to the VCA arrives from the VCA fader, VCA groups, cut switch, and automation computer. Figure 13.3 illustrates this.
Digital fader
A digital fader simply determines a coefficient value by which samples are multiplied. For example, doubling the sample value—a coefficient of 2—results in approximately 6 dB boost. Halving the sample value—a coefficient of 0.5—results in approximately 6 dB attenuation. Other coefficient values can be easily calculated based on a simple formula.
Figure 13.3 A single VCA can be used with many level-related functions.
Scales
Figure 13.4 shows a typical fader scale. There are perhaps as many fader scales as manufacturers and software developers. While the actual numbers might be different, the principles discussed in this section are common to the majority of faders.
The typical scale unit is dB, which bears a strong relationship to the way our ears perceive loudness. Most often, the various steps are based on either 10 dB (which is subjectively a doubling or halving of perceived loudness) or 6 dB (which is approximately doubling or halving of voltage or sample value). The 0 dB point is also known as unity gain— a position at which a fader neither boosts nor attenuates the signal level. Any position above 0 dB denotes a level boost, while anything below denotes attenuation. Most faders can boost; therefore, they provide an extra-gain. Common extra-gain figures are 6, 10, and 12 dB. We should regard this extra-gain as an emergency range—using it implies non-optimal gain structure and possible (often minor) degradations to the sound. Ideally, this extra-gain should not be used, except on rare occasions when all the mix levels are set and a specific track still calls for a bit of push above unity gain. At the very bottom of the scale, there is –∞ (minus infinity), a position at which the signal is inaudible (often achieved by attenuation of 120 dB or more, but some analog desks cut the signal when the fader reaches this position).
One very important thing that becomes evident from Figure 13.4 is that, although the different scale steps are evenly distributed, the dB gaps between them are not consistent. At the top of the scale (between –10 dB and +10 dB), the steps are 5 dB apart; at the bottom (–30 to –70 dB), the steps are 20 dB apart. This means that sliding the fader at its lower end will cause more drastic level changes than sliding it at its higher end. Another way to look at this is that at the top end of the fader level, the changes are more precise— clearly something we want while mixing. This suggests that we want to have our faders in the high-resolution area of the scale; provided we do not use any extra-gain, the area between –20 and 0 dB is the most critical. One useful function found on some audio sequencers is the ability to key in the required level or have more precise control over the fader position using modifier keys.
Figure 13.4 A typical fader scale.
Figure 13.5 A demonstration of the uneven fader scale. These eight Pro Tools faders are set at –10 dB intervals. This demonstrates how hard it can be to set any value within the –70 to –60 dB range, while it is quite easy within the –20 to 0 dB range. Modifier keys can accompany mouse movements for more precise results.
Another consequence of the uneven scale is that, when we gain-ride a fader, the lower it goes, the more drastic the attenuation becomes (and the opposite effect for raising). If we fade out a track, for example, we might have to slow down as we approach the bottom of the scale.
Working with faders
Level planning
A common question probably asked by every mixing engineer at least once is: So which fader goes up first and where exactly does it go? There are a few things we have to consider in order to answer this question.
Faders like to go up. It should not come as a surprise if throughout a mixing session one by one all the faders go up—a few times—ending up at exactly the same relative positions. This is the outcome of the louder-perceived-better axiom. Here is an example of how easily things can go wrong; you listen to your mix and find yourself unsure about the level of the snare. You then boost the level of the snare. This is likely to make it more defined, so the move appears to be right. A few moments after, when working on the vocals, you realize that they do not stand out since the snare is nearly as loud, so you boost the vocals. The vocals should now stand out all right. Then, you find that the kick seems too weak since the snare is way louder. So you boost the kick. But now you are missing some drum ambiance so you boost the overhead, then the toms, then the cymbals, then the bass, then the guitars, and before you know it you are back to square one—you are unsure about the snare again.
Faders like to go up.
There are a few ways to solve this cyclical syndrome. First, understand that faders are the least-sophisticated tools for making something stand out. Second, listening in mix-perspective minimizes the likelihood of individual level moves. Third, it takes some discipline to stick to the plan, especially when it comes to level boosts. After absorbing these three ideas, you might still find that faders like to go up, so the ultimate solution is this:
Leave some extra-gain available.
The extra-gain dead-end is shown in Figure 13.6. It happens when a fader is fully up, but the instrument can still use some gain. This scenario can be solved in various ways, but it would be much easier planning the levels right in the first place, so the extra-gain deadend never happens. We have said already that the fader’s extra-gain is better kept for an emergency. So we normally do not want any faders above 0 dB at the early mixing stage. If we also take into account that faders like to go up, it might be wise to leave even more additional gain, so there is still some distance to go before we start using the extra-gain. For example, it might be wise to start the mix with the loudest instrument set to –6 dB (which gives us 6 dB of virtual extra-gain on top of the standard extra-gain).
Level planning requires setting the loudest instrument of the mix first, and then the rest of the faders in relation to it. An example of the opening mix moves for a production where the lead vocal is expected to be the loudest would involve setting the lead vocal track to –6 dB and then setting the overheads level with respect to the lead vocal. The vocal track can then be muted if one wishes to start mixing from the drums. Rough mixes can help as well—if the rough mix ends with the highest fader at +4 dB, that fader should open the real mix at 0 dB (or slightly below it), with the rest of the mix then built in comparison
Figure 13.6 The extra-gain dead-end. The vocal track is at the top position of+12 dB, which means that no additional level boost can be applied by the fader if needed.
to this level. Another strategy is to create a quick balance mix (faders only) of key tracks before mixing onset. Then adjust the fader levels so the highest fader is on or just below 0 dB.
So that is the digital version of level planning. Analog mixing works slightly differently, and we can borrow the analog wisdom for the benefit of digital. Unlike software sequencers, analog desks do not have a hard clipping threshold—levels can go above 0 dB. When planning levels on analog, engineers sometimes look at the VU meters—a far better indication of loudness than the peak meters on a software mixer. One might mix a dance track in Logic and start by setting the kick to –6 dB. Then, when another instrument is introduced, it might need some extra-gain, and can even end up at the fader’s dead-end. Employing a VU meter during the early stages of level planning can be beneficial in digital all the same.
The extremes-inward experiment
Sometimes level decisions are hard to make and we find it difficult to ascertain how loud a specific instrument should be. The extremes-inward experiment can help in these situations. Accompanied by Figure 13.7, the process is as follows:
- Take the fader all the way down.
- Bring it up gradually until the level seems reasonable.
- Mark the fader position.
- Take the fader all the way up (or to a point where the instrument is clearly too loud).
- Bring it down gradually until the level seems reasonable.
- Mark the fader position.
- You should now have two marks that set the limits of a level window. Now set the instrument level within this window based on the importance of the instrument.
If the result of the experiment is that the window is too wide, say more than 6 dB, it suggests that some compression or equalization might be beneficial.
Figure 13.7 An illustration of the extremes-inward experiment.