REW question about measurements for sub implementation

[john61ct]

speaker building perspective where the colocated subs and mains become a "full range speaker"

Yes what I am also shooting for.

With additional subs in mono away from front LRC for room mode smoothing and when "wider LP" is desired.

 

[john61ct]

Look at it this way: moving subs up/down has the same effect on vertical modes as moving the sub left/right and front/back on horizontal modes. It's just another axis in x, y, z. There is nothing special about the vertical axis. If a dip in the response is caused by a front-to-back mode, elevating the sub will not do very much.

Yes the Z axis is commonly discussed (sub crawl) I just haven't come across Y that much.

I guess "keep subs on the floor" being a default for practicality, convenience, most being big & heavy.

 

[thcdru2k]

I think it helps to separate four different things that often get lumped together: placement, room interaction, sub to sub integration, and main to sub integration. A colocated or stacked sub and main can behave more like a single full range source, which can be advantageous for crossover integration. The tradeoff is that the sub and main are exciting the room from nearly the same location, so you generally lose some of the modal averaging benefits that come from using multiple source locations. The room response is therefore determined not only by placement, but also by the phase relationship between the left main and sub, the right main and sub, and ultimately the interaction between the left and right bass sources themselves. Good results require looking at all of these separately because a problem can originate from the room, from the integration between a sub and its main, or from phase interactions between the two left and right sub/main systems.

 

[thcdru2k]

Tryng to align four acoustic sources through a crossover region while each source is interacting with different boundary conditions, distances, and room modes. The odds that all of those phase relationships naturally line up over a broad frequency range are fairly low.

The more sources you have contributing through the same frequency region, the more likely there is some beneficial phase rotation available

 

[john61ct]

In the mono sub arena, that "more overlapping the better" is apparently very helpful in MSO's approach.

I beieve there are some commonalities with MIMO tools like Gsonic Immersive and Dirac ART.

Are there others?

 

[Daverich4]

I am currently writing a subwoofer integration guide for Acourate. Although it is for Acourate, the general principles apply no matter what DSP software you use. You can view the draft here.

This is a very broad overview of the steps required:

Step 1: Place your subwoofer. DON'T avoid the sub crawl, you want to fix as many problems as possible BEFORE you begin the DSP process. You only have to do the sub crawl once! Don't be lazy! You spent money on this thing, you want to get the best out of it, don't you?
Step 2: Decide on your LPF/HPF config. See the draft eBook for considerations on how to decide.
Step 3: Time and phase alignment. See the draft eBook for the method. Instead of using Acourate's tool, use REW's alignment tool.
Step 4: Do a MMM or multi-point averaged measurement and EQ that to flat.

Re: 75dB measurement volume. Don't bother, just measure at your normal listening volume. Loudspeakers and subs are nonlinear devices, if you measure at a different volume you may get a different frequency response. You only need a 75dB measurement if you are doing formal testing.

I’m planning on adding two subwoofers so I’m not quite sure how the subwoofer process works in that case. Do it do it with one sub first and get that setup and then do it again with the second subwoofer?

 

[thcdru2k]

Most people will advocate getting the subs working together first and then treating them as a single bass system before integrating with the mains. I personally do not as I do not think there is necessarily a right or wrong way.

I think the most important thing is not the exact order, but that you measure the individual sources as well as the various combinations. Looking at the individual measurements tells you what each source is contributing, while the combined measurements tell you how they are interacting. Without both, it can be difficult to determine whether a peak, dip, or cancellation is coming from the room, a particular subwoofer, sub to sub interaction, or main to sub integration.

The combined measurement is ultimately what matters, but the individual measurements are often what explain why the combined measurement looks the way it does.

 

[Keith_W]

I’m planning on adding two subwoofers so I’m not quite sure how the subwoofer process works in that case. Do it do it with one sub first and get that setup and then do it again with the second subwoofer?

It's covered in the guide. For mono subs:

- Measure L+R speakers together. Call this L+R.
- Align SUB1 to L+R. Call this SUB1+L+R.
- Align SUB2 to SUB1+L+R. Call this SUB2+SUB1+L+R.
- Continue until all subs are aligned.

 

[Daverich4]

It's covered in the guide. For mono subs:

- Measure L+R speakers together. Call this L+R.
- Align SUB1 to L+R. Call this SUB1+L+R.
- Align SUB2 to SUB1+L+R. Call this SUB2+SUB1+L+R.
- Continue until all subs are aligned.

My question actually was about finding the best location for them before doing all that. If there was only one sub I would use the sub crawl to find the best location and then folllow the process you’ve described in the e-book that I have. But how does that work when there’s more than one sub that needs to be located?

 

[Keith_W]

My question actually was about finding the best location for them before doing all that. If there was only one sub I would use the sub crawl to find the best location and then folllow the process you’ve described in the e-book that I have. But how does that work when there’s more than one sub that needs to be located?

I am going to cowardly duck that question and tell you what I did when I helped my friend set up the system. We chose the two candidate positions and placed the subs there. There was a dip that we could not get rid of, so we put the sub in another position and experimented again with timing to see if we could get rid of the null. It worked.

I can't come up with a solution for every scenario that will work the first time. Sometimes you just have to experiment and iterate until you get it right.

 

[Daverich4]

I am going to cowardly duck that question and tell you what I did when I helped my friend set up the system. We chose the two candidate positions and placed the subs there. There was a dip that we could not get rid of, so we put the sub in another position and experimented again with timing to see if we could get rid of the null. It worked.

I can't come up with a solution for every scenario that will work the first time. Sometimes you just have to experiment and iterate until you get it right.

Thanks for the response, you've been really helpful a number of times. As have both of your eBooks.