System Tuning Best Practice questions

Hi Folks

As I continue to be impressed by the expertise and workflows of my fellow engineers in the field, I am eager to learn about the best industry practices that you have found effective when tuning a system. And admittedly, I just may be getting confused on some basic principles, so…there's that. haha!

1.) The Pro Audio Workshop course is structured as EQ, then phase align Sub/Main. I've noticed that some engineers align first and then EQ to a target trace. Does this sequence matter? What is your preferred workflow for system tuning and EQ'ing to Target traces?

2.) When aligning phase slopes, I know to use only one side of the Mains. What if there are multiple subs? Let's say 5 subs in an array across the front of the stage. Do I use all the subs and one side of the main? Or just the subs on the side closest to the mains I am using? (For example, using L Main and all subs? Or L Main and two subs closest to LMain? Etc.). How do you handle this situation?

3.) Finally, when I use SMAART to EQ, do I use only one side of the system (as I would for phase alignment), or is the whole system turned on?

As always, your help as I tackle this is very much appreciated. Feel free to point me to community discussions or modules specifically addressing these topics. I could not find any that would answer these specific questions through an initial search.
Thanks,
Eric

Hi Eric!

PSA: Use whatever method gets results that work. I feel like I have to say this every time because otherwise people insist that, "X woks for me every time!" If it X works for you, use X.

Does this sequence matter?

Yes. You can't skip physics, but you can skip steps if you are confident in their results.

Here's an example workflow:

1. EQ and level set A solo.
2. EQ and level set B solo.
3. Align A to B.
4. EQ A+B together.

You can skip the first two steps in you are comfortable with simply EQing the combined result afterwards. For example, I often jump straight to alignment for main+sub, knowing that I'll just EQ the combined result afterwards. If I had plenty of time, I probably wouldn't do that, though.

When people skip solo EQ, I assume, it's because they want to save time and want to prioritize the combined result. Main+sub is a good example because there is no place in the audience where you won't hear the combined result, so why not just work with the combined result from the beginning if it saves time?

When aligning phase slopes, I know to use only one side of the Mains.

This is a preference. Technically accurate is align with both sides on. But of course, no one does that in field because it's too difficult.

Let's say 5 subs in an array across the front of the stage.

Should I solo one side of my sub array during alignment?

do I use only one side of the system (as I would for phase alignment), or is the whole system turned on?

Similar to the previous discussion, every source need solo EQ and combined EQ, but you can skip solo EQ and just straight to combined EQ if you need/want to. At this point in my life, I skip a lot of steps. Assuming that each source has some place in the audience where it is in isolation (which is not the case for delay speakers), this is what I do:

1. Align everything using SubAligner.
2. Turn the entire system on. Take 8 measurements for each source in its coverage area, skipping symmetrical sources.

Then I come back to my computer and handle the EQ. This gets most of the way and I do the sound check or the show. If there's more time, I will go back and verify some alignment points, maybe do more EQ, but it's mostly refinements at that point.

Thanks for the explanation.

If I use separate matrices for subs and tops, once I align them, any EQ changes (or Level changes for that matter) have to be made to the combination, and can no longer be made to the individual matrices. I have to get the amplitudes as close to the target curve as possible before alignment. Correct?

In other words, post-alignment EQ changes need to be made on the LR bus (or, a group that combines Subs and Tops or something similar). If I make individual changes to the separate matices (or level changes), I risk changing the phase relationship.

Have I understood that correctly?

once I align them, any EQ changes (or Level changes for that matter) have to be made to the combination

That is the easy way, yes, but of course, if you know what you're doing, you can always go back and edit the solo settings. This is why it's a good idea to do some investigation ahead of time to know what your limits are. For example, on the Balance Levels page of SubAligner you can see the maximum recommended attenuation and addition that you can make before getting into risky territory with the alignment. If you do an alignment between a Q1 and Q-Sub, for example, the maximum recommended gain of the sub is +3dB. This means that if you raise the sub level by 3dB that the phase response through the xovr region will still be within 60º.

Also, this might help:
The EQ Puzzle: What filters will break my perfect subwoofer crossover alignment?

and can no longer be made to the individual matrices.

You could do it with a DCA, but yes.

I have to get the amplitudes as close to the target curve as possible before alignment.

Right. You need to set EQ and level first to know the area of interest. The goal with a unity crossover is that the sum will hit the target. With an overlap crossover you typically just EQ the result to hit the target.

post-alignment EQ changes need to be made on the LR bus (or, a group that combines Subs and Tops or something similar).

Yes, something upstream. In your DSP you would have an upstream EQ block or put those outputs into a group somehow.

Have I understood that correctly?

Yes, I'd just caution you to not obsess about it too much. Try to stay flexible in your workflow so that if you need to go back and change something you can do that. It is common to finish alignment and then someone decides that the that several speakers must move.

Good work on this Eric!