Set Your Audio Recording Levels Using Sensible Gain Staging
Recording And Gain Staging
The old adage of recording signals as hot as possible really only applies to the analog domain, not the digital one. With analog reording this was because of the noise floor inherent in analog circuitry and tape, where you were fighting low level noise and hiss. This was also how we were taught to record when digital recording was relatively new, when the best we had was 16-bit audio. Back them everyone was trying to record as close as possible to 0 dB without clipping. However, now with 24-bit converters there is no need to record signals nearly so hot. You can get great recordings with the levels set much lower, and then you will also find it MUCH EASIER to mix later, because you are not trying to force so many loud signals into the headroom limitations of a DAW. If you have a number of tracks all recorded very loud, and approaching 0 dB, at mixdown you will find yourself running out of headroom, and will have to bring all your track levels down in order to compensate.
The professional +4 standard reference point of 0 dB in the analog domain ( 1.23 volts) is actually at -20 from 0 dBFS (top of the Digital scale) in your DAW meters. So with the reference level of an analog console being +4 (unity Gain or “O”) and equalling -20 in the digital world of your DAW, if you slam everything up too hot and close to zero in your software mixer, it will be like trying to jam all your signals on a console at maximum level into the master section. Not such a good idea. If we run a signal into the line input of a high-end console at a level where the channel meter reads 0 vu,( +4 or 1.23 volts) it will be at unity gain. A great console might be able to handle up to about +20 or even +24dB, which means we have about 16-20 dB of headroom above the 0 VU point before the signal turns to custard. Keeping our recording levels lower makes it much easier to mix things later. By the same token then, if we have -20 dBFS as our “0” reference in our DAW, we then have 20 dB of headroom before we clip the signal. Clippng a digital signal sounds far worse than clipping an analog one, so steering well clear of that should be our aim. Digital clipping does not sound good. If you look at a clipped waveform up close, you will see it is square at the point of clipping. There isn’t any logical reason to record with levels even close to approaching the level where they clip. It certainly won’t improve the sound of a recording, and you run the risk of having little headroom left for mixing.
The dynamic range of 24-bit recording is theoretically about 144dB. With 16-bit it is only 96dB. With the dynamic range being 144 dB, and if the noise floor is about 120 dB, then the effective dynamic range or signal to noise ratio is 120 dB. That is plenty of room for recording with levels well below 0 dB before you introduce unwanted noise into the signal.
So, during the recording process trying to aim for an average of between -15 to -10 dB on the meters of the DAW tracks, with peaks not going above -6 dB is a good level. However, it is perfectly fine to record at -20 or even lower.
Mixing And Gain Staging
Effective use of gain staging when recording sets you up for a good mix. Keeping your gain staging at sensible levels during mixing is the secret to obtaining clean, clear, transparent mixes with plenty of headroom. Proper gain staging will not reduce punch in a mix. If each track is recorded at a level in your DAW which is not pushing them too hard, and you send various instruments to Busses or sub groups, you will find that you don’t clip the inputs of any plugin compressors or EQ’s that you place on the busses, and you have every opportunity to boost the outputs of those pugins or busses to make up for any perceived lack of punch you may have because your levels are lower. You can then have some cool punchy buss sub mixes, that can then be summed to your main stereo buss, without digitally distorting anything anywhere in the chain.
If you find that signals in your DAW have been recorded too hot for clean mixing with plenty of headroom, the use of trim plugins is the first place to go. Using one of these as your first plugin on each channel, you can adjust the level going into EQ’s, compressors and other processors. This is not the same as just pulling down the fader, because that only affects the signal level going to the master buss or summing busses. The trim plugin acts like the gain control on your preamp, whereas the fader is like the output volume control. Another advantage of this is that if you insert a hardware compressor or EQ into a channel it will be receiving a signal around the 1.23V (+4) that it was designed to operate at.
Headroom is the term used to describe the difference between the maximum level that can be achieved without clipping a signal, and the average level of the signal.
Definition: Gain Staging
The term “Gain Staging” refers to the level of a signal beginning at its source, and ending at its final destination. Along the signal path, you may have several points where changes to the signal level can be made. Proper gain staging refers to setting sensible gain levels at these various possible adjustment points. This gives us our gain structure. An example of gain structure would be a vocalist who can sing louder or quieter, going into a microphone, then into a preamp, followed by a compressor, and then into the A/D converter to record. All points in that signal chain are places where the gain could be changed.
Reference Line Level
-20 dBFS = 0 dBVU = 1.23V
By Tony Koretz
© copyright March 2012