Thursday, March 9, 2017

Understanding Gain-Staging

If you don't understand what gain-staging is, it refers to the process that audio engineers and producers and DJ's deal with in order to ensure that when they're playing audio through a system with several different places to change the volume, that the final signal that comes out of the system is as clean and strong as possible, without exceeding the "maximum" 0 dB level that marks the beginning of distorted audio signal.

Putting this another way (from Wikipedia):  "Gain staging is the process of managing the relative levels in a series of gain stages to prevent introduction of noise and distortion.  Ideal gain staging occurs when each component in an audio signal flow is receiving and transmitting signal in the optimum region of its dynamic range."




The 0 dB level confuses many people.  That sounds like it should be a level with absolutely no signal, not a level with a high amount of signal.  Well, that's a long and complicated topic, so you'll just have to trust me on this for now.  Think of 0 dB as being the highest "good" signal strength, and signals that are weaker than that (or "quieter") have negative numbers going down as they get even quieter.  If you want a detailed explanation of that, and have some time, you can watch this video that I produced:



In some systems, moderate amount of distortion isn't bad.  A bit of warm distortion on a nice crunchy electric guitar can sound pretty good.  But in other situations, especially when dealing with a digital audio signal, distortion can sound bad almost immediately.

Imagine this situation that a DJ might face:
- Sound signal comes out of a piece of line level equipment such as a pitch-controlled CD player.  Or it might alternatively come out of a turntable, which will mean that the signal will need to be boosted by a pre-amp before being processed through a mixer or mixing console (usually, this is built into the console).
- Entering the "back" of the mixing console, the usual path for signal flow is "top to bottom" or "back to front" (although not always).
- The signal coming in from the equipment goes through a "gain" or "trim" knob on the channel strip.
- The signal then gets affected by the position of the Volume Fader on the channel strip.
- The signal then leaves the channel strip and goes to the Master Bus (often just called master).  The signal might get borrowed on the way by booth monitors, headphones, etc.
- On the master, the signal can be amplified (increased) or attenuated (decreased) by the master volume fader.
- The signal then leaves the mixer and often goes to an amplifier, which has another volume control.
- The signal leaves the amps and goes to the speakers.

Incidentally, sometimes the signal leaving the mixer will go straight to a speaker, because some speakers are called "powered speakers" and have their own amplifiers built right in to the speaker.

At any place that the signal is "low," there's a greater amount of background/system noise being introduced to the signal, RELATIVE to the overall strength of the signal.  This is called the Signal-To-Noise ratio.  Higher numbers are useful, because you want a greater amount of signal compared to the background noise level.

Of course, if the signal is too high and exceeds the 0 dB level, distortion can be introduced, so we try to avoid that.

Although different engineers and producers and DJ's and musicians will have different views on the matter, which can legitimately vary depending on the work that they're doing and the system they're using, a good general rule is that a signal should be moderately close to 0 dB at its peaks, but to still leave a gap.  This gap is called the Headroom.

If I'm producing music, I often try to have the smallest headroom gap (the peaks of the audio signal) to be several decibels (dB) below the 0 dB mark, but the valleys where the signal is weaker can be much lower than that.  So my average signal strength may end up being possibly at -12 dB or -10 dB. It varies.

If I'm DJ'ing and paying attention to the signal levels flowing through my system, the same approach is valid.  My peaks will be close to 0 dB (which is sometimes indicated by a red light on the VU level meters of the equipment).  My average will be a bit below that, but not too low.

As your signal passes through each of the various volume points in the system, it's great to have them all consistently treated roughly the same.  I'd rather see a lot of "strong but not too strong" signals at different points in the system, rather than some extremely weak (low/quiet) and some extremely strong (hot/high/loud).  The end result may be the same in both cases, but in the system which has a lot of "too weak" and "too strong," there's going to be more background noise AND more likelihood of distortion.  It's like Goldilocks and the Three Bears.  You don't want your porridge to be too hot or too cold ... you want it to be "just right."


Ok, now that you have a basic understanding, here are links to several pages which will give you more information:

  Sweetwater:
www.sweetwater.com/insync/gain-staging

  Sound On Sound:
www.soundonsound.com/techniques/gain-staging-your-daw-software

  Home Studio Center:
www.homestudiocenter.com/gain-staging

  Gearslutz:
www.gearslutz.com/board/rap-hip-hop-engineering-production/589213-proper-gain-staging-thread.html



And finally, here a video about Gain Staging that seems to have gotten a lot of positive comments:




Good luck with your DJ'ing, audio setups, and production projects!

Jonathan Clark (DJ Bolivia)
www.djbolivia.ca


For a complete list of my own tutorial videos about music, DJ'ing, production, audio engineering, and recording, visit: