What About Aux-Fed Subs?
A full exploration of this topic can go on for quite some time, but there’s one related concept that deserves a mention – a common objection to the aux-fed technique is that varying the level of the subwoofer drive (or the contribution of individual inputs) will move the acoustic crossover point. I touched on this in the previous article, so interested readers are encouraged to review that first. (Go ahead — I’ll wait here.)
One way to avoid these shenanigans is to use the sub send as a fixed-level group – each input is either routed there, or it’s not. But what if we vary the level? Does the acoustic crossover point move? Yes, it does. But how much?
We used a low shelf filter to attenuate the low end of the main PA response by 10 dB below 500 Hz, done in the pursuit of a flatter overall response but for the sake of this examination, it has the same effect on the acoustic crossover point as raising the level of the sub drive by 10 dB.
Figure 3 shows that this moves our acoustic crossover point to around 125 Hz, at which point our phase alignment is just barely holding together. This isn’t always the case.
Figure 4 provides data from another large-format rig, in which the alignment is maintained out past 250 Hz. No realistic amount of level increase on the part of the front of house engineer would result in pushing the subs and mains out of phase alignment here.
In conclusion, although there are some real, measurable issues that we need to be aware of when it comes to sub alignment and aux-fed subwoofers, I certainly don’t feel that these are show-ruiners, and in the case of aux-fed subs, some engineers (including myself) may happily make that trade for the benefits of the AFS mixing workflow – cleaning up the low end of the mix by controlling which inputs can drive the subs.
Of course, everyone has their own approach, but based on this data, I think it’s fair to say that fear of lurking phase gremlins shouldn’t dissuade folks who might otherwise be interested in trying the approach.
The author would like to thank David Williams and Hannah Goodine for their contributions to this article.