What happens if the system has an analog crossover that doesn’t allow separating the high-pass and low-pass points?
This is fairly common, especially in powered loudspeakers with built-in processing.
In this case, you’d reach for the EQ instead of applying more gain to the sub. A low-shelf filter will boost (or cut) everything under the corner frequency.
Since this EQ would be upstream of the crossover it affects the subs and mids equally, but will not change the relationship between the two.
Therefore, the crossover overlap doesn’t change, so there won’t be problems with multiple drivers reproducing the same frequencies.
Tuning the sub and low-mid suggests a track with a good walking bass line. Listen as the bass moves up and down the scale, and determine if there are spots where the bass drops out or gets louder.
One favorite is “Sailing to Philadelphia” by Mark Knopfler and James Taylor, which offers plenty of bass both below and above 100 Hz. It helps in setting the level of the subs in relation to the mids by listening to the bass line as it moves through the crossover region.
Don’t make the mistake of relying on just one song – use several in different musical keys. For example, if your test music is in the key of F, you may be missing some room resonances that are excited by music in Eb.
Back To Tuning
Now that we have relative levels between the subs and low-mids, we can time align the crossover point. This can be done by flipping one band out of polarity and playing a pure sine wave right at the crossover frequency. (e.g., if the crossover point is 100 Hz, use a 100 Hz tone.)
Then delay one passband until there is the maximum amount of attenuation of that 100 Hz tone. Perfect cancellation rarely happens from two different sources like this, but you can find the spot where the most cancellation occurs. This method zeros in on the proper delay time between two adjacent passbands.
Always use one side of the PA – it’s the only way to hear the point of maximum attenuation. Just remember to copy the delay time to the other stack(s), if the processor doesn’t do it for you.
The “gotchas” to watch for are starting with the wrong polarity, and/or delaying the wrong passband. If both passbands are in polarity with each other, flip one band in and out of polarity to determine if the crossover frequency gets louder or softer. I never know for sure what the absolute polarity is, but that’s a minor issue as long as the two passbands are both playing together on the same team.
Delay the passband in the drivers closest to you. With front-loaded subs and mids, there may not be need for delay. However, sometimes the mids will be located on the rear of the subs to make the stack more physically stable. In this case, delay the subs since they’re closer to you.
This method can be applied to the rest of the passbands, although it gets harder to do by ear as frequencies go higher. Applying delay to a high-frequency horn driver is difficult without measurement tools; however, an educated guess can be made based on the physical depth of the driver in relation to the other drivers in the system.
For example, if the HF horn is 12 inches deep and you’re trying to delay a 10-inch mid driver in relation to it, we know the mid has a depth of about 4 inches from the gasket to the voice coil, and thus the delay time has to be about 8 inches (12-inch horn depth minus 4-inch mid depth). This works out to be about .6 milliseconds (ms).
I first apply .3 ms worth of delay, work up to about .8 ms, and observe where the maximum attenuation occurs. It’s in there somewhere, but due to the physics of the smaller wavelengths, we may not be able to get it perfect by using our ears alone.
One caveat: if the mains are flown with the subs on the ground/deck, there’s no exact way to get the time alignment perfect because every position in the venue will require a different delay time. Find a happy medium and go with it.
Flown PA with ground-stacked subs is a great thing, but it means that the sub-to-mid delay time will be a compromise.