The importance of accurate engineering
Audio engineer Aston Fearon offers his advice on how to ensure the highest levels of precision when mixing a live show.
Audio engineer and technician Aston Fearon offers his advice on how to ensure the highest levels of precision when tuning a system or mixing a live show.
In our pursuit of exceptional quality sound for our audiences it’s important not only to think about the tools we have at our disposal but also the accuracy of our techniques.
We can achieve good results using rough estimations and making coarse changes but the more accurate we can be with our changes, the more robust the systems we use will become (with more than adequate headroom) and the more we can get each instrument to fit neatly where we want it to in the mix. This might mean avoiding certain tools and techniques as well as realising the limitations of human judgement to produce the precision we need.
The problem with graphic EQs is that because the bands are fixed they are unable to ring out a room very accurately or efficiently. They come close and can get the job done but they often take other frequencies, which aren’t a problem, down with them. If the offending frequency in question is 575Hz, for example, I would have to pull a notch out of 500Hz and 630Hz to try to achieve the desired notch, which is less than ideal because the notch at 500Hz and 630Hz would evidently be more severe than if I could pull a notch out of 575Hz. Parametric EQ is usually much more accurate because of its sweepable nature and Q.
Finding the central frequencies that are likely to ring in a room is important but if we don’t seek to see how many frequencies either side of it are also a problem then we won’t be able to define the Q of our EQ bandwidth appropriately. Too narrow and we aren’t being thorough enough; too wide and we are destroying too much of the signal in the same way that using a graphic does when we need to pull out two adjacent frequencies.
Real-time analysis (RTA)
Being able to name a frequency by ear is a good skill for a sound engineer to have but because graphic equalisers only have 31 bands, the accuracy at which we can name that frequency is either limited by these 31 bands or difficult to be exact because of the number of actual frequencies in the spectrum of human hearing. There is often debate about whether to use our ears only or also incorporate visual and numerical elements into our engineering processes. Personally I think the answer is both. My ears might tell me to take a notch out of 10kHz but Rational Acoustics’ SMAART might identify the frequency as 9.5kHz. I still need to verify whether I agree with SMAART’s analysis but it can give me a lot of useful information to help me make decisions more accurately.
My ears can tell when the combination of two or more sources affects its tone undesirably. Although I know this is being caused by identical sound waves which arrive at slightly different times within a cycle, my ears can notice the effects but cannot accurately hear phase itself. It’s far more effective for me to use SMAART to distinguish how much the sources are out of phase and where I need to move my boxes or add delay to get an accurate alignment. A wavefront that is phase incoherent will negate accurate EQ choice further upfield at the desk, even if a source sounds great in the headphones.
Using a reference like a good pair of headphones (or nearfield monitors) can help us to listen to our sound sources before they hit the PA. If we are happy with what we hear but it doesn’t sound the same through the PA then we can use system EQ to correct any discrepancies in the system rather than trying to fix the problem at the channel EQ. If we solve the problem at the channel level rather than the system level we’ll have to correct multiple channels and when we use the desk file again on a well-tuned system the channel EQ will be void.
Once we know our system is tuned accurately we can also be sure that anything we PFL (Pre-Fade Listen) will be the same as what the audience will hear. Line checking in this way is only easy if we have an accurate reference point to use. The more accurate this reference point (hopefully with little coloration) and the more familiar we are with how it sounds; the more accurate we can be when we make tonal changes to individual signals.
All of these considerations can give us an extra degree of resolution here and there, which may seem negligible, but added together the results can create a big difference to the clarity of a mix and the efficiency and headroom of a system – both at the amplification stage and the SPL we mix at. This means a system can work less to produce a result that sounds even better to the ears of our audience.
Aston Fearon is a freelance sound engineer and technician mostly working at the FOH end of the multi.