There’s a running joke in the system optimization community: “How much time does it take to tune a sound system?” The answer (“as much time as you have”) is often all too true – tuning sessions can devolve into hours of minute tweaking, over thinking, over processing.
But that need not be the case – effective and efficient tuning sessions are all about broad strokes. Of course, the common truisms apply here: if you get the design, loudspeaker placements and aiming correct, the alignment should be fast and simple. “Design to align.” The first 20 percent of the tuning time gets you 80 percent of the way there – and so forth.
How does this actually translate in practice? What tips, strategies, and workflows can we use to work our way through a complete system alignment in 20 minutes rather than two hours?
Let’s step through my process for a typical alignment in a 1,700-seat theater. It (hopefully) goes without saying that there are multiple avenues by which to remove the skin from this conceptual feline, so feel free to steal the helpful bits and leave the rest. Our focus here is on the broad strokes – the major signposts along the way.
An efficient alignment session starts with a good design, since there’s no magical EQ settings that will compensate for a loudspeaker placed in the wrong spot or aimed the wrong way – mechanical issues need to be resolved mechanically.
Basis Of Expectations
My process starts by examining the design in prediction software, adjusting placement, trim height and splay angles until we land as closely as possible to the coverage consistency we’re aiming for. Most prediction platforms allow the user to view the expected frequency response at various locations in the audience plane – this data forms the basis of my expectations for what we should see once the system is hung and powered on in the space. If they don’t match, that will be a sign that something’s wrong.
Once the system is up in the air, we begin the boring but all-important process of verification. Skip this step at your own peril as it’s your “pre-flight check” to reveal system issues.
We start by “spraying” pink noise through each system zone. First, the left main array and then right array – and make sure they match. If they don’t match, stop and find out why. Then we move into each array zone: left top, left middle, left bottom, to make sure that each is correctly wired and patched. Then check the right array zone by zone, each sub zone, and front fills.
If anything comes up in the wrong spot or far louder or quieter than expected, we stop and find out why. This typically takes less than a minute (five percent of our 20-minute goal).
Now that we’ve verified that everything is behaving as it should, let’s tune this thing. Since we’ve verified the system is symmetrical (left matches right), we can save ourselves time by tuning the left half, copying those settings to the right side when we’re done, and then checking that symmetry is preserved.
We start by prioritizing the “papa bear” systems that cover the largest portion of the audience (mains) and work our way down to the smallest (front fills). With the main array, we’re looking for overall tonal curve as well as variance over the coverage area, and will “kill both of these birds” at once by measuring on axis to each of the array zones (typically back, middle, and front of coverage area).
Overall tonality is set with EQ on the entire array, and HF can be massaged per zone for increased consistency. If the mains are single point source boxes, our EQ decision must be “one size fits all” so focus on the common trends that show up at all the measurement positions – those are the ones that can be chased with a high likelihood of success.
If subwoofers are split left/right, then next let’s bring on the left side of the subs and set their level to achieve the desired LF extension and “haystack.” This is a good time to set main-sub timing – put some thought into where you want them to be in time – because they can’t be in time everywhere.
A combined measurement (mains+subs) at this point should establish the desired tonality of the system and will serve as our target for all the other zones. Due to a desire to restrict the amount of unnecessary LF energy being dumped into the room, supporting systems will get high-passed to varying degrees such that they only supply the needed portion of the spectrum.
For example, let’s tackle the front fills. Front fills fill a spatial gap (hence the name) in both level and frequency. With the mains on, find the “gap” where the front fills will need to do the heavy lifting – typically the first few rows. A measurement there with the mains on and the front fills muted will reveal the shortage of direct HF energy but likely more than enough LF energy. Although there are multiple ways to approach front fill tuning, my approach with front fills will fill that spectral hole – and no more.
Since there’s no shortage of 100 Hz in the front rows, the front fills don’t need to supply it. This keeps that energy from washing the stage and also keeps everything cleaner and tighter. Taken together, the front fills restore the first few rows to our target response.
Now, for the time, find the “seam” – the area where the front fills and mains meet at equal level – and set the time there. That’s where they’re going to be most interactive, so that’s where their time relationship is most critical.
Even in venues with thousands of seats, these decisions happen at very specific points: tonal and level decisions happen on-axis to the zones in question, and then timing happens at the seams where our systems overlap. Speed and efficiency comes largely from having a clear game plan – knowing exactly where the mic is heading next and what decision you’re going to make when you get there.
Surgeons move quickly and efficiently not by rushing but by eliminating unnecessary movements. In a large venue, a good chunk of tuning time can be spent walking back and forth between the mic and the computer (analyzer and control software). Multiple microphones and measurement-quality wireless microphone systems really earn their keep in these larger rooms.
In addition, a screenshare or Client Window functionality for your analyzer software means that the measurement and control work can happen on a tablet or laptop carried with you rather than running back to front of house each time you need to capture a measurement or adjust a DSP parameter.
Why It Matters
Another important ingredient here is expectations. As noted earlier – does the system do what the prediction said it would? By extension – do the measurement results make sense? Do they agree with what we perceive with our hearing and with our sight?
By getting into the habit of forming a mental expectation before taking each measurement, you build in an excellent “error detector” and a check against making bad decisions based on faulty data or assumptions.
The final element is experience and practice. How the system looks, how it sounds, and how it measures all dump into the big mental stew of “eye-ear” coordination and reinforce each other over time. If something measures fine but sounds weird, find out why. If something sounds great but measures weird, find out why.
All these tools at our disposal bring more information to bear, which give us better context on our situation and thus are better equipped to make a good decision. Like anything else, it gets better the more you do it.