Any good technology system with multiple components goes through a basic creation process, whether that is your home stereo system, an IT network or a professional AV system. First, you design the system—you select the proper components, determine how they should be connected and identify their proper setup so they can accomplish the goal the system is intended to solve. Then, you install the system—you physically mount and connect all of the pieces of the system together.
Finally, you configure the system—this involves adjusting all the settings and performing any tasks that take the system from “physically ready” to “performance-ready.” For an audio system, this involves (among other things) room optimization, which includes adjusting EQs, delays and other settings to account for unique aspects of a room (reflections, materials that absorb different parts of the audio spectrum, etc.) so the system will sound the best it can.
Although there are other aspects that come into play with room optimization (delay, etc.), we’ll focus on the primary one, which is equalization. Now, this is not the equalization of a single instrument or vocal mic that may be adjusted depending on the particular event. Instead, we are talking about more basic equalization settings that are set once when the room is configured and then often don’t need to be adjusted again unless something in the room is changed that affects the sound of the room or sound system itself.
When we think about EQ, there are three basic types. First, there is the channel EQ on the soundboard, which is the EQ you may think of when I say equalization and the EQ the sound engineer will adjust for an event. The second EQ you will encounter is the speaker EQ. This ensures the transducer sounds its best within the speaker cabinet. These days this is often set within the digital signal processor (DSP) built into the amplifier and the correct setting is typically provided by the speaker manufacturer. Crown amps, for example, have multiple built-in DSPs and contain preset speaker EQ settings for various JBL speakers. Finally, there is the room EQ. This is where you can correct issues caused by the way the sound interacts with the physical room itself. This is the EQ that we are most concerned about in this article.
There are a number of ways to go about adjusting room EQ, and opinions about which method is the “best” are just as numerous. That said, I’ll list a few of the popular methods that are available and the process. The first method is perhaps the least scientific, but still remains popular and can be effective for engineers with well-trained ears. This is what I’ll call the “mix to your ears” method. In this process, you play a familiar recording through the sound system (clean with no channel EQ adjustments or effects) and then adjust the room EQ until the recording sounds “right” to you. There is, of course, some debate on what constitutes a “good” test recording (I tend to prefer “Another Brick in the Wall Part 2” by Pink Floyd for this), but the important thing is that the engineer be extremely familiar with how the recording “should” sound and has a lot of experience. Of course, you can also monitor through a good set of flat-response reference headphones for comparison.
The second process is what is known as “ringing out” a sound system. This method attempts to identify feedback issues and echoes that effect sound balance, and involves connecting a clean microphone and then cancelling out feedback on any infringing frequencies. You start by taking each vocal or instrument microphone and plugging it into a clean channel on the board. Turn the gain to negative infinity (off), bring the master output fader to unity (0dB), and then set the fader for the microphone channel to +5db. Slowly turn up the sound until there is just a bit of feedback when you say “check.”
For a stereo installation, you will want to use the pan to EQ each side separately. While saying your “mic check” phrases (I like the “Speak the Speech” monologue in Hamlet III.ii), take a channel (or a node with a narrow Q, if parametric) on the EQ and slowly turn up the gain for the channel/node. Slowly test each channel on the graphic EQ (or adjust the node frequency). When the feedback increases and you hear either a “ringing”/feedback or a hollow/booming sound, you have found an offending frequency. Turn down the gain so the “peak” becomes a “scoop” and the sound should not feedback anymore at that frequency. Repeat for other channels/nodes until there is no more feedback.
Of course, many argue that this is still not that scientific (though proponents point to its effectiveness). However, there are other methods as well. For example, some dbx DriveRack signal processors have AutoEQ™, an algorithm that provides a fast, accurate and automatic EQ of a room. The processor plays sine waves in particular frequency sweeps. An attached Real-Time Analyzer (or RTA) reference microphone “listens” to the room as the tones play, and the processor automatically adjusts the EQ in a matter of seconds.
Finally, you can also adjust the system using room optimization software or a hardware spectrum analyzer. In this approach, the tools use anywhere from two to eight or more microphones to provide detailed spatial analysis of the entire space. You can then adjust the EQ for groups of speakers (a single array, for example). You then copy and adjust across the space until you have as consistent a sound as possible throughout the entire room. This process is obviously the most detailed and can require training to do properly (SynAudCon is a great place to start), but it also provides the sound consistency that top engineers say is vital to a great experience.
The particular approach you choose will greatly depend on your particular situation. Which approach do you prefer (and for what type of application)? Let us know in the comments.