Gain structure is one of the most important topics in audio, but I don’t think it gets the attention it deserves. A solid understanding of gain structure is absolutely essential for doing work with audio.
The principles in this post apply to every facet of audio – from mixing music to optimizing the performance of a sound system.
I’ve got some really helpful tips for you, but if this is our first time meeting, my name is Kyle. Welcome to Audio University.
What Is Gain Staging?
First of all, “gain” refers to the level difference between the input and output of a device. Applying +6 dB of gain to a signal will make it 6 dB louder than before.
Proper gain staging or gain structure means that the level of the signal is optimized throughout the signal chain.
If the level of the signal isn’t optimized at any particular stage, the quality of the signal will be degraded.
Sometimes the degradation of the signal is obvious, and other times it’s more subtle. But even subtle problems can add up to become bigger problems, so it’s very important that you keep these simple principles in mind.
What Is Good Gain Staging?
As audio technology has changed over the years, the way we approach gain structure has also changed. However, the goals have mostly remained the same – to minimize noise and distortion while maximizing dynamic range and headroom.
Let’s take a closer look at how gain structure can help us to achieve these goals…
Clipping, Overdrive, & Saturation
I’m sure you’ve heard the effect of turning up an audio system beyond its limitations. As the signal begins to exceed the limits of the system, the signal will be distorted.
Now, don’t get me wrong – this isn’t always a bad thing. Muddy Waters changed the course of music with his innovative guitar sound by harnessing what some might call “improper gain staging”!
However, different types of systems will distort the signal in different ways. For example, the effect of overdriving the input of an analog-to-digital converter is far less pleasing than the effect of overdriving the preamp tubes of a guitar amplifier.
This is one of the biggest differences between analog and digital equipment. The consequences of overdriving an analog circuit or analog tape are usually more forgiving than the consequences of overdriving a digital system.
While saturating analog tape might result in a rich and warm tone, clipping a digital system sounds terrible almost every time.
In cases where you want a clean recording, you should try to keep the level of the audio signal within the limitations of the audio system.
It isn’t just the individual channels you need to worry about, either. You may have encountered a situation where no individual channels are clipping, but the master meter is solid red.
It’s best to leave some headroom in your mixing system so that even the combined level of all of the inputs won’t exceed the limitations of the master bus. Later on, I’ll show you how to set up a mix to make sure that you get this right.
For right now, it’s just important that you understand the importance of leaving headroom in your system so that your audio signal won’t be distorted.
However, overdrive and distortion aren’t the only considerations when it comes to setting up gain structure… We also want to structure the gain of a system to minimize noise.
Signal-to-Noise Ratio & Dynamic Range
The signal-to-noise ratio describes the difference between the signal and the noise floor. Generally speaking, the quieter the noise can be in comparison to the signal, the better.
Let’s say you want to record the sound of your voice with a tape machine.
Using the gain knob on your microphone preamp, you’re able to control the level of the signal to be recorded. Turn the knob up, you’ll get a louder recording – turn the knob down, you’ll get a quieter recording.
To avoid overdriving the tape machine, you’re careful not to turn the signal up too loud. But you should also consider what might happen if the signal is too quiet.
The tape has a certain level of inherent noise, or hiss. The tape noise stays the same no matter how high or low you set your mic preamp.
If you record a very weak signal to the tape, the signal-to-noise ratio will be very small.
Plus, once your voice is recorded, you will need to turn up the signal to bring it to a usable level. When you boost the signal, you’ll then be boosting the noise as well.
If a strong signal is sent to the tape, the signal-to-noise ratio will be much greater.
And it’s not just analog tape that has inherent noise. All devices will add some amount of noise to the signals passing through them.
Each time you add another device to the signal chain, the signal will acquire more noise. It may not seem like a substantial problem until you hear the combined noise floor of all of those devices.
As the noise floor rises, the quietest signals will be indistinguishable from the noise. For example, the gradual decay of reverb will be much longer with a low noise floor than with a high noise floor.
This brings up another big difference between analog and digital audio – the noise floor.
While you’ve got to be very careful not to record too weak a signal to analog tape or through analog devices, the noise floor of digital systems is much lower and therefore much more forgiving.
Setting Audio Levels
When setting levels, you’ll be searching for the sweet spot between too high (which could lead to distortion) and too low (which will shrink the signal-to-noise ratio).
Like I said before, the approach is different between analog and digital equipment.
With analog equipment, we tend to set the levels a bit higher because the noise is our biggest concern. Slightly exceeding the limits of the analog circuit may even sound desirable.
With digital systems, we tend to set the levels a bit lower, because there is a very low noise floor and the sound of digital clipping can completely ruin a recording.
VU Meters (0 dBVU) vs Full Scale Meters (-18 dBFS)
You’ll even see a difference between the type of meters used in analog vs digital systems.
When setting levels in the analog realm, we will usually aim for peaks around 0 dBVU on a VU meter which is equal to +4 dBu. +4 dBu is the line level standard in professional audio systems.
Most digital systems use an entirely different scale for metering, called Full Scale.
When setting levels in a digital system, the peaks should never exceed 0 dB Full Scale, or 0 dBFS. If a signal reaches 0 dBFS, it won’t be properly quantized by the system, and you’ll start to hear the sound of digital clipping.
Opinions on this vary, but I like to set input levels so that the meter averages around -18 dBFS and the peaks don’t exceed -12 dBFS. This gives the signal lots of headroom while still paying attention to maximizing the signal-to-noise ratio.
Gain Staging: How To Control Gain
Throughout a signal chain, there are many places where gain can be affected. It’s important to be aware of each stage of gain and set each of them to optimize the level of the audio signal.
Let’s take a look at some common places in a signal chain where gain can be affected.
Device Volume Level
The first place you can control the gain of a signal is the output volume of the audio source.
I usually set the output level of smartphones, laptops, DJ turntables, and keyboards to about 80%. Outputting a healthy level like this means I won’t have to increase the level of the signal as much down the line.
Preamp
The second stage is usually the input preamp.
Like I mentioned before, I will usually set the preamp so that the input meter shows the peaks at about -12 dB Full Scale in a digital system or 0 dBVU in an analog system.
Region Gain
In your DAW software, you can sometimes adjust the gain of a region.
This is a good place to set initial levels for your mix if you’re mixing tracks that were sent to you at a level that is too high or too low.
Filters & EQ
When using an equalizer, you’re adjusting the gain of the signal.
Boosting the high frequencies with a high shelf will increase the overall level of the signal. Cutting low frequencies with a high pass filter will decrease the overall level.
Dynamics
You should also consider the effects of dynamics processing, like compressors, expanders, and duckers.
These tools will often have a make-up gain knob that allows you to increase the level to compensate for the gain reduction you’ve applied.
Aux Sends, Faders, & Masters
Each auxiliary send knob, channel fader, bus master, and master fader is also a place where you can influence the gain structure of your signal chain.
In general, it’s best to start with all faders at unity gain so that signals pass through without being boosted or attenuated.
It’s also important to ensure that you aren’t clipping the input or output of any auxiliary sends or FX returns.
Amplifier Volume
Rather than using the master fader as a system output volume knob, it’s best to keep the master fader at unity and adjust the loudness of the system with the monitor knob or the amplifier volume knob to find an appropriate level in the room.
This will help to ensure you are working within the ideal range of your mixing system.
Gain Staging: Setting Up A Mix
Let me show you how I would go about setting up a mix within a session in my DAW. These steps might seem simple, but this is probably where 90% of mixes go wrong.
I want to set the initial level of the tracks so that they are at a workable level that still leaves some headroom in the master bus.
In some DAWs, you can adjust the region gain directly. If your DAW doesn’t have this feature, you can simply drop in a Trim or Gain plugin on each channel.
I’m going to set the level of each track so that the meter on each input is at about -12 dBFS. If you do this, you’ll notice that there is still plenty of headroom on the master meter with all track faders at unity.
Taking the time to set up the mix in this way will allow you to work comfortably within the limits of the system.
Plus, some plugins (especially those that model vintage gear) are designed to work within this range in order to provide the best, most authentic sound and experience.