System Tuning 101

One of the first tasks that I’ve undertaken as I dive headfirst into audio at New Life Church is the tuning of our PA systems in our main rooms.  However, before addressing the journey of each room from where it began to where it has ended up, I thought it would be good to outline as concisely as I can a philosophy for how I think a PA should be configured and what equalizers in the chain should be used for what purpose.  Eventually, over the course of multiple posts, I hope to come around full circle to not only show what I’m trying to accomplish in the room, but how that translates to an audio for video capture process as well.

In my mind there are three places in a system where equalizers normally reside.  CHANNEL EQ –> SYSTEM EQ –> SYSTEM PROCESSOR. We’re going to talk about them out of this order in order to get my point across so here we go…

First is the channel EQ that exists on the mixing console.  I view the responsibility of this part of the chain to provide correction or artistic shaping of the interaction between a microphone and an instrument or vocal.  If a mic is too woofy on a vocal or a guitar amp is too harsh (and you’ve already tried to improve the situation at the source by working with the artist), the channel EQ is the place to make the necessary adjustments.  Carrying this out, once you’ve made all of the corrections to the various inputs, you should then be able to listen to the results of this mix through headphones or record the left/right mix and play it back on another speaker system and have a good representation of what you intended the mix to sound like.  It shouldn’t be too bright or too dull because you’ve fixed all of those individual combinations on the channel strips.

The second equalizer/processing location in a large sound system is the system processor which addresses the PA speakers themselves – the crossover points of a bi or tri-amped system, signal delay necessary to get all of the speakers arriving in a common time domain, and equalization that addresses the natural interactions between all of the speakers in an overall system.  Normally this degree of processing occurs inside a system processor – a Dolby Lake, BSS London, proprietary amplifier, or in my case, Shure P4800′s.  It is located in the chain directly before the signal coming from the sound mixer hits the amplifiers that drive the various speakers.

The final equalizer/processing location is the overall system equalizer.  This point in the chain falls between the output of the mixer and the input to the system processor/amplifier/speakers.  Sometimes this processing point might be combined into the system processor, in the old days it used to be simply a trusty 31 band stereo graphic equalizer.  The purpose of this equalization step is to optimize and correct the speaker system as a whole to the acoustics of the room where it is located.  If the room is boomy or harsh, live or dead, etc. the system EQ allows the engineer to craft the overall sound of the speakers so that what comes out of the mixer sounds transparent (meaning what came out of the mixer is what comes out of the speakers) in the room.

I outlined the three parts in the order I did because I kind of look at them in that order of mental processing as well.  The channel EQ corrects for individual inputs and builds the mix itself, the system processor corrects for all of the individual speakers and amplifiers and builds them into a cohesive system, and then the system EQ is the glue that brings those first two pieces together into what is heard in the room.  If any of those three places in the chain is not properly optimized, the overall product will suffer.

The primary goal I’m trying to achieve as I think about these three parts of the sound system is transparency.  I want what sound came out of the console to come out of the speakers with the same sonic characteristics (transparency).  The simplest way to test this is to play a known recorded track (I have a playlist of songs that I use to test how a PA sounds) through the speakers and see if what comes out sounds like how I know that track should sound because I’ve listened to it thousands of times through hundreds of different PA’s – the balance of low end to top end, the harshness or lack thereoff, the warmth of the system, etc.  Why does this matter?  Because if the PA is not transparent, if what comes in is NOT what goes out on the other end, I’m going to have to make corrections somewhere else in the chain (normally at the same places on every channel’s EQ) to make the mix work and this will likely compromise that board mix.  It also makes it far more difficult for a more novice engineer to achieve a great mix unless they are more comfortable with channel EQs or for me to get something great happening quickly because I’m going to have to apply more corrective EQ which will take more time to dial in.

In my experience, the biggest weakness I find in installed PA systems is in the optimization of the system EQ.  More often than not (and exactly what I’ve dealt with at NLC.tv), the system EQ does not make the PA system transparent.  Playing an iPod track through the console without any channel EQ will sound far different coming out of the speakers than it would listening through good headphones on the console or a different speaker system.  The core sonic characteristics of the song do not translate through the room.  As a result, the engineer needs to apply channel EQ to correct for these room problems in order to start mixing with a blank canvas.  In our iPod example this can be easily done because we’re only talking about a single stereo input.  However, explode this out to a 6 or 7 piece band, 5-8 vocals, and 30 or more channels and the task can be far more daunting.  The byproduct is that the L/R board mix, which often feeds video recording or internet feeds, suffers greatly.  All of the board mix changes that are made to make inputs work in the poorly tuned room are more drastic than what was really needed so the L/R mix isolated by itself is nothing to be excited about.

Next time we’ll start getting specific about a tuning philosophy to correct these problems.