1. The only good reason for uncoupled subs.
2. When to point your subs at the wall.
3. Time saving weapon.
4. The only reliable shelf prediction?

The only good reason for uncoupled subs.

While teaching Sub School I came across an interesting fact: The only time I split my subwoofers up to cover the left and right side of the audience is when the width is greater than the depth.

The beginning of forward aspect ratio (FAR) is 1:1, a square. If your audience is a square then you need a 180º speaker, or in this case, a sub array. I find this is pretty common, which is one of the reasons that the gradient array is so useful.

I only know a handful of coverage angles that you can hit with gradient array variations, though. Outside of that handful, I need a sub arc. A sub arc comes with other requirements, one of which is that it needs at least six elements.

Today I only have two subs. In this case you’ve got to break up the array or risk under exposing the sides of the audience.

This is also making me realize that someone should Subwoofer Array Calculator for all of the gradient array variations.

When to point your subs at the wall.

Have you ever noticed subs placed near a wall will have a weird dip in the response? Or maybe you just noticed that subs indoors suck in general?

Another guideline I follow is that if the subs are going to be near a wall then they should just couple with it. How near? I think the 3:1 rule works here. If the reflection path is less than 3x the direct path, you’ll have significant ripple.

Today the subs are going to be almost up against the wall anway, so all I had to do was turn them around to face the wall. If you want to be precise, open up the LFACC calculator to estimate the acoustic center of the sub. Just be ready to answer questions from curious project managers.

Someone asked for an explanation of direct path. In this case, it’s the shortest path from the wave front origin to the listener. Here’s an image comparing a reflection from a wall with a mirror image phantom source on the other side of the wall and the resulting ripple cased by their summation. I tried to draw the sound coming out of the subwoofer and then curving a bit to get around the box itself. Hopefully that’s a bit more realistic.

I have an alignment secret weapon and so can you.

Some days like today I just want to skip alignment. It’s a small show. There’s a million other things to do. Who would know?

Unfortunately, I would know. And then all during the show I’d be wondering, “Am I hearing a problem with phase, time, or something else?”

It was one of those days where there’s too much to do and not enough time (or people!). (aka just another Wednesday)

There was me, the video tech, two hands and seven hours to build the sound, video, and lighting systems, which might not sound like a lot if everything came in preconfigured racks, but this av company likes to keep all hardware separate for rental purposes. It might take half an hour just to wire up all of the RF receivers correctly, you know?!

I thought about skipping it.

Then I remembered, I have a secret weapon. SubAligner calculates all of that for me.

Instead of my previous many step process of system alignment where I would use reference docs and calculators to find the best alignment position, gather reference data for comparison, setup my audio analyzer, capture data, run various calculations to handle the noisy data…

…now I just fill out the form in SubAligner.

(of course, I love playing with audio analyzers, so at the end of the day I took some extra time to set it up and verify my results)

My next fear: Is this weird old Bose subwoofer even in SubAligner?

Yes, it is!

Why?

Because I went to the av company’s warehouse three years ago and measured everything in their inventory. Now I’m confident that any gig I work on for them will be covered.

I can help you do the same thing.

Take a day out of your life to measure all of your speakers and you’ll never need to worry about alignment again. I’ll even get on Zoom to walk you through it. Just send me a message.

The show the next day went great! When the first music cue started I was really happy I took the time to work on the alignment.

The only reliable shelf prediction?

Have you noticed that every mixing console seems to handle the low and high shelving filters differently?

I work on a lot of small and medium sized corporate events with no real output processing. Everyone does it in the console. I hate it, but I have never found a tiny output processor that I can just throw in my pelican, so here we are.

Today I’m mixing on an SQ-7, one of the worst consoles for output processing. Otherwise, it’s fine for mixing. They’re just not people to abuse it like we do. All of the filter widths are set with bandwidth, which is fine, just not what I’m expecting. There are only four and the bandwidth is pretty limited.

If I insert a 2.3oct wide filter at 1kHz with a gain of -6dB, then I am expecting the see the stop band at -6dB and the cutoff frequency somewhere near half the shelving filter gain. Instead I get this.

The SQ-7 stopband is at -5.5dB and the cutoff is at -5dB. If I want to make these filters match I have to change my low shelf filter in CrossLite to 2.3kHz, -4.3dB, 2.3bw.

I’m not so concerned with who is right and who is wrong. I just want a reliable way to do this work.

Luckily, I always setup my audio analyzer to use the console output as the reference input to my audio analyzer. This makes switching over to take a measurement of the console itself pretty easy. All I need to do is switch my audio analyzer reference to the audio interface’s internal loopback and the mic input to the console. That’s how I got the SQ-7 measurement above and that’s how I confirmed that the the low-shelf I set in CrossLite would produce the same results in the SQ-7.

Note: A few days later, at a CrossLite meeting, Rafael Pereira gave me a great suggestion. If I’m already measuring the console in real time, then I can just look at the sum between the filter measurement and my speaker measurement against the target for a faster more accurate prediction of the combined result. Can’t wait to try it on another show! 🙂

What do you think?