Gain structure — setting levels through the signal chain

What gain structure is

A recording or live sound system is a chain of devices: microphone, preamplifier, analogue-to-digital converter, digital audio workstation, digital-to-analogue converter, amplifier, loudspeaker. Each device has an optimal operating window — a range of input levels where it performs best, sitting above its noise floor and below its clip point.

Gain structure is the discipline of setting the gain at each stage so that the signal stays within that window throughout the chain. A well-structured gain chain minimises noise, avoids distortion, and gives the system its maximum usable dynamic range.

The problem with poor gain structure

Too much gain early, too little later

If excessive gain is applied at the microphone preamplifier and then attenuated further down the chain, the noise generated in early stages is amplified along with the signal. This noise cannot be removed downstream — once it is in the signal, it stays there.

Too little gain early, too much later

Conversely, if the preamplifier is set too conservatively and gain is recovered later, the signal sits too close to the converter's noise floor during the critical recording stage. Every subsequent gain increase also raises the noise that was recorded.

The principle

Apply the right amount of gain at the right stage. Front-load enough gain to raise the signal comfortably above the noise floor of each device, before passing it to the next stage.

Unity gain

Unity gain means a stage passes signal through with no change in level: the output equals the input. A gain of 1×, or 0 dB. In an ideal signal chain, every stage after the initial gain-setting stage would operate at unity gain, with level changes made deliberately and intentionally.

In practice, mixing involves level adjustments throughout the chain, but the concept of unity gain is useful as a reference point. A fader at unity (often marked 0 or U on a console or DAW) means you are passing signal through that stage without adding or removing level.

Setting gain at the preamplifier

The preamplifier is the most critical gain stage in a recording chain. The goal is to:

1. Raise the microphone's output (typically −60 to −40 dBu for a dynamic mic on a normal source) to line level (+4 dBu nominal)
2. Do so without overloading the preamp's input stage on transients

A common approach is to set the gain so that loud peaks on the source read approximately −18 dBFS on the converter's meter, leaving 18 dB of headroom for unexpected transients. This aligns the nominal signal with the analogue equivalent of +4 dBu on a properly aligned converter.

Gain staging in the DAW

Inside a digital audio workstation, the noise floor of floating-point arithmetic is negligible. However, gain staging still matters for two reasons:

Summing headroom: When many tracks are combined, their levels add. If every track is running near 0 dBFS, the sum will clip. Keeping individual tracks at conservatively low levels — peaks around −18 to −12 dBFS — leaves room for the mix bus to accumulate without overloading.

Plugin headroom: Many audio plugins, particularly those modelling analogue equipment, have an internal operating level at which they are designed to work. Running a vintage compressor plugin at 0 dBFS may push it into unintended saturation behaviour, or miss the saturation character entirely if levels are too low.

Gain vs fader

In most console and DAW workflows, two controls affect level: a gain (or trim) control, and a fader. These serve different functions:

• Gain sets the input level — it is a fixed calibration of how loud the source is relative to the system's nominal level. It should be set once and not adjusted during a mix.
• Fader is a dynamic control used to balance levels in the mix. It rides on top of a correctly set gain.

Reaching for the fader to compensate for a poorly set gain is a common mistake that compromises headroom and noise performance.

A practical workflow

1. Set preamplifier gain so that the loudest expected moments peak at around −18 dBFS on the converter's input meter.
2. Trim clips in the DAW so that individual track peaks sit at a consistent level (−18 to −12 dBFS is a common target).
3. Use faders to balance the mix, not to compensate for large level differences between tracks.
4. Monitor the mix bus — if the master fader is peaking above −6 dBFS before limiting or mastering, individual track levels are likely too high.
5. Apply appropriate output gain at the final stage (master bus processing, export limiting) to reach the target delivery level.

Headroom for mastering

Mixes intended for mastering should be delivered with headroom — typically peaking at −6 dBFS or lower. A mastering engineer will adjust overall level as part of the mastering process. Delivering a mix that is already limited to 0 dBFS removes their ability to work with the dynamics.