In thinking about the church sound systems I’ve worked on over the course of my career, I realized that the vast majority of the problems I’ve encountered can be boiled down into just a few categories. If a church asks me to address a problem, odds are that it was caused by one of the following issues.
1. Gain Structure
This is of paramount importance in an audio system, to the point that if it’s wrong, it almost doesn’t matter if everything else is right. In simple terms, every component of a sound system is designed to operate with signals of a certain level. Too high, and the signal becomes distorted when the device overloads; too low, and the signal becomes swamped in the noise floor.
We generally classify audio devices as operating at one of three signal levels: mic level (obvious…), speaker level (also obvious) and line level (everything else). Mics put out a very weak signal, on the magnitude of a hundredth of a volt or so, simply because the air pressure variations they’re picking up are not very strong to begin with. So as soon as the mic-level signals get to the console, they hit the preamp, which boosts them up to a more reasonable level.
From this point on, we’re operating at or around line level, which in the professional audio world is +4 dBu (1.23 volts) or thereabouts. We hang out at line level all the way until the signal leaves the power amplifiers, which boost both voltage and current significantly. We’re talking about levels of power that can rival what comes out of the receptacles in your home. This is appropriately named speaker level, and should be treated with respect.
Within the realm of line-level signal flow, which is where our audio signals spend most of their electronic existence, it’s extremely important to pay close attention to how the signal level changes as it flows through the desk.
Every piece of audio equipment known to humanity has a maximum signal level it can accommodate (usually around 10 volts rms for pro devices) and a noise floor which, as long as we occupy this physical universe we do, can only be reduced to a point.
If the signal level is too high at any point in the chain, distortion may be added to the signal as a result of overload and we’re stuck with it for good, regardless of what we do downstream. A common example: let’s say we set the mic preamp gain too high, and the signal is clipped as it comes in. We can lower the channel fader to attenuate the signal back down to a reasonable level, but now it’s just quieter and still distorted.
Likewise, if the preamp is set too low, the signal-to-noise ratio (SNR) will be poor, and the signal will be noisy and hissy. We can boost the signal later on, but we’ll be boosting the hiss, too. So it’s very important to get the signal level right at every point, in order to avoid causing problems that we can’t reverse later.
Optimally, every device in the signal chain would overload at the same time, meaning there’s no headroom bottlenecks anywhere. If the console meters are at -10 dBu (most analog desk meters read in dBu, even if they don’t say so) and the amps are clipping, that’s no good. Likewise, if we’re slamming all the faders up to the top of their tracks and we still can’t get a decent level into the amplifier, that’s a problem as well. The goal is to get the signal level into the “sweet spot” (engineer-speak: “nominal level”) as soon as possible and keep it there.
How is this done? Here’s a popular method:
• Set the output faders (that means bus masters, such as the main L/R mix) at unity, or 0 dB. This means the fader circuit is not changing the gain of the signal at all, just passing it all on through.
• Set the input faders up to unity (again, move along, nothing to see here).
• (Slowly) raise the input gain until the desired level is attained.
If feedback sets in before there’s enough level (that’s why I said “slowly”), check mic placement and pickup pattern, check monitor placement, and/or grab a graphic EQ and reduce the offending frequency. With more attention paid to gain structure, feedback should happen less often.
What’s been accomplished?
The signal is flowing through the console at the optimal level, adding just the right amount of gain. Let’s say we followed the old “analog tape” style of gain structure, where the gain is cranked up until the signal lights up at a predetermined level on the input meter, and then only raise the fader as much as needed.
This practice originated in the recording studio, where it was important to record to analog tape at a certain level for the best recording. (Too high, distortion. Too low, hissy tape noise. Sound familiar?) Engineers would crank the gain to get the signal onto the tape at the optimum level, and then upon playback lower the fader to get the track to the desired level in the mix.
In the live paradigm, we’re not trying to beat tape hiss. By cranking the preamp and keeping the fader low, all we’re accomplishing is an amplification followed by an attenuation. This is not optimal because, as we’ve now learned, this creates a part of the signal chain that’s going to overload before the rest of it, so the entire system just lost headroom. (As an added annoyance, there would also be two different gain structures for the pre-fader and post-fader auxiliary sends, which just complicates things further.) Starting with everything “up” and then adding only as much gain as needed makes it much simpler to understand a signal’s gain structure, since we’re not adding extra stages of amplification and attenuation.
When adding effects, either inserted (like a compressor) or on a send-return loop (like reverb) the same rules apply. Start with input and output controls set to unity and adjust only if warranted. If you’re sending a bunch of inputs to a reverb, you may be lighting up the unit’s input meters enough to warrant turning it down some. If the compressor is creating more than a few dB of gain reduction, try adding some back with the makeup main control. Again, we’re trying to keep the signal from getting too high or too low at any point.
The exception to this “open it up full and let ‘er rip” approach is power amplifiers. Those knobs on the front of the amp? They’re not volume controls. Think of them more like sensitivity controls: they set how much voltage is required to drive the amp to full power. This is a common misconception.
Turning down the amp inputs does not mean that the amp can no longer reach full power, it just means that it takes a higher voltage to get there. Ideally, if the amps were properly matched to the loudspeakers, this control can be adjusted without worrying about the loudspeakers blowing up. Caution: if you’re not absolutely positive about this, don’t touch it.
During the mix, if the console meters are at a healthy level but the amp inputs show more than occasional clipping, consider turning down the amps because they’re causing the bottleneck in the gain structure. Conversely, things might seem a bit on the quiet side, but don’t raise the amp’s input levels unless you’ve verified that you’re not going to torch your loudspeakers by doing so.