Using two (or four) identical subs in your room, of adequate output to achieve your goals***, gives you higher quality, more consistent and more accurate bass.
In almost all instances, the quality and accuracy of the bass is better when you split your sub budget in half and get two subs.
Set all your speakers to "small."
Send the subwoofer signal out of your AVR or pre-pro as a single mono signal from a single output using a Y splitter.**
Place the subs (and seats) in the most favorable positions.
Align the subs with the mains (often called the “distance tweak" since it is accomplished via the distance setting on an AVR.)
Select the crossover points for your speakers.
Run room EQ. (If your room EQ system changes your distance setting for the sub, change it back to what you calculated in Step 4.)
First, one needs to place the subs (and seats) in the most favorable positions.
The concept is that you place the subs as best you can for your given room.
If you have multiple rows of seats, you likely want to place one sub in each corner of a rectangular room. You might be able to get by with two subs, likely placed in the center of the front and rear wall.
If you have only one row of seats, you can usually manage with just two subs and they can both go on the front wall like you see in marketing photos if:
you have some placement flexibility on the front wall, and most importantly,
you have the ability to move the seating forward or back to avoid dips along the length of the room.
Often, putting those two subs at 1/4 and 3/4 of the front wall or of back wall (both in front or both behind) will give you good left/right consistency across seats. It just won’t deal with having good bass at every distance from the screen (ie multiple rows) nearly as well as the centers of the front and rear walls.
Todd Welti did some great research at Harman on subwoofer placement, which gives you a visual reference for what locations tend to work the best in symmetrical rooms, so if that is what you have, this can get your started. .https://www.harman.com/documents/multsubs_0.pdf
Some subwoofer placement options give you a big head start in terms of getting consistent bass response across multiple seats, typically with fewer dips in response to deal with. Those diagrams give you a head start.
Where to place the seats: If you have the choice, and you have a rectangular room, you can estimate the places most likely to have the best (as in least bad) bass response in the room, so it is worth starting at these locations, if you can. Even if you can place your seats in these places that are likely to be the least bad, there will still be significant problems with your bass response, hence this guide to using multiple subwoofers to dial in your sound more accurately. Anthony Grimani, in his CEDIA training, likes to use this diagram showing which locations are the least problematic.
Of course, rooms are not acoustically symmetrical even when they measure that way with a tape measure. Different construction, door jambs, etc, make them acoustically not quite symmetrical especially in the bass.
So all of these guidelines need to be checked in the real world, with a sine wave generator, to see where peaks and dips actually are…. and then to confirm the efficacy of your selected subwoofer locations, and seating locations.
You can use any of the online tools for predicting what your room modes are. Then, using those frequencies as a starting point, check where dips (nulls) occur by listening to tones of those frequencies generated by a single subwoofer.
Than, switch on that second sub. If the symmetry is adequate, then will no longer hear most of those modes…… and when you do they will, at least, be about the same at every seat, and mostly just peaks, not nulls. Why is that good? Because if it is the same at every seat, you can EQ it and it will sound good at every seat!
With many modern AVR’s, you may be able to stop at this point and let the AVR do the rest of the setup, distance setting, crossovers and EQ work!
OPTIONAL: You can time align (phase align) the subs with one another.
This creates a single virtual subwoofer, with positive summation (ie, fewer dips and better peaks) by manipulating the subs independently. The challenge, here, is that doing this for one seat is relatively simple. Doing it for multiple seats is a gargantuan process, best left to software like MSO (Multi Sub Optimizer) and the use of a miniDSP. In a mostly symmetrical room, where you can hear via sine wave test tones that nulls are largely cancelled, this is not necessary. For the same of completeness, I’ll describe the manual process for a single seat with two subs:
Level or gain match then with an SPL. Yes there is a difference between these two methods. But either will work.
Then time align: Leave one sub's phase at zero. And adjust the other subs phase until the combined response of the two subs produces the fewest troughs in response at your main seat.
Here is an example where alignment was necessary, because I did not place the subs symmetrically in the room:
I thought I was going to have to bust out the miniDSP for my subs and that would let me adjust the distance even more than the phase knob, but then I remembered that the phase knob on the Rythmik is also a distance knob (even tells you happily what 180 degrees is in milliseconds of delay), and found via the REW alignment tool that about 7ms was what I needed.
Green and Purple are the individual subs.
Red is the combined response without time alignment.
Yellow is the combined response with time alignment.
For simplicity, I do not show all the other traces that didn't look as good as the yellow version.
Pretty darn text book summation where the combined response meets or exceeds (especially in the critical chest slam 40 to 60hz region) the response of any individual sub by quite a margin, in many cases by more than 10db.
Note that I tried to use the REW time alignment prediction tool. It wasn't quite right. So it pays to take time to experiment with different "distances" or "phase."
If you have more than two subs, once you have dialed in two subs, consider those a single, virtual sub, and then align that single virtual sub with the third sub, and so on, for as many subs as you have.
If you have three or more subs, it can be worthwhile to use a miniDSP and to use the free Multi Sub Optimizer software. This video shows the process with a miniDSP. Personally, I think you can likely do the job with simple phase knob adjustments on many systems. So feel free to use this approach but to adjust “delay” with the phase knob on your sub(s) before going the extra step to use a miniDSP:
Finally......check your other seats. If you have significantly different response in your different seats, you will want to consider using MSO to help dial this in. In a non symmetrical room, or a system with compromised subwoofer placement options, or with vastly differing subs, the MSO approach may be required.*
Figure out the AVR distance and crossover settings.
These two steps actually interact with one another, so you will work on them iteratively.
One needs to align this single virtual subwoofer (ie, all the time aligned subs as a single sub output) with the mains. In the old days we called this phase alignment, but the normal tool here is the “distance” setting on the AVR.
This was popularized as the sub distance tweak.
To get this right is an iterative process where you measure the response in that region with different delays (distances) for the virtual single sub, and with different crossover points.
You can see quickly that that means measuring every distance from zero to 30ms, with a 60hz crossover, then all those distances again with a 70hz crossover, etc, up to 120hz or even higher. There will likely be a combination that is a little better at showing no wide dips……and that is the best alignment.
Don’t worry about peaks. You will EQ the final result.
And remember: Sometimes the best crossover may be, for example, a 120hz crossover, and that may feel weird…… and sometimes you may do better at 80hz……..etc. It largely depends on room modes, and whether the mains are positioned better to cancel them out than the subwoofers are (hint: not usually).
Personally, since that lower frequency choice also increases seat to seat variance in the 80 to xxx region, I find in such a situation that the 120hz crossover in this example might actually be the better compromise. Better alignment with the mains and better (less) seat to seat variation due to the advantage of multiple subs covering more of the frequency range.
The verification will be the REW measurements of the final result showing a positive summation throughout the crossover region. And while it should go without saying, your auto setup software may not choose the best distance nor the best crossover for you. It doesn't actually test the speakers and subs TOGETHER nor with different real world settings. It measures them individually and tries to predict what will work best.
Here is an example of the concepts at work in a real room:
I started with 80hz and checked the distance from 10' down to 0' which is more than a 180 degree phase shift in the crossover region, so should cover most options.
2' of distance was the best compromise. Some were a tiny bit better in the 80hz region, some were a little better in the high 90s, but those involved tradeoffs in one or the other area. 2' was the best balance.
Here is a closeup showing the region a bit better:
Then I tested different crossovers to see what could be gained....And 80hz was still the winner.
Then I tested where things broke down when using the Center speaker instead of the Right speaker as the bed layer speaker I'm integrating the subwoofer with. No worries:
Finally I tested the surround speaker crossover. Going lower than ARC chose (100hz) did create a better graph, but the power handling of the Triad Bronze satellite when trying to play at 90hz was not up to the task. Its a small specialized speaker and just a temporary choice. Bumping the crossover up to 120hz didn't make it worse measuring than when at 100hz, and increased the power handling, so that is what I settled on for this speaker for now. (I'll be eager, short term, to reclaim my on wall Revels for this room, and longer term to use the Triad Bronze LCR for surround duty when I build out the false falls.)
Always remember that the measurement tools can make these differences look really big. But the truth is that audibly, just getting "in the ballpark" will often work very well. That is why the automated systems, even if they aren't as thorough as this iterative process of testing real configuration changes, often can sound good.
What do I mean?
Many of these glaring differences disappear when human hearing is taken into account.....and arguably if one completely trusts psychoacoustic smoothing, 4' might be a better distance setting for the sub than 2' by a tiny margin (ie, 1db in some places). Or, in other words, I have graphed the measurements to exaggerate as much as possible what the microphone can pick up and what the graphs can show, but if you take into account what a human can hear, things are much less stark:
Then run your room EQ, which will clean things up super well.
You can also add in a house curve if you like, with something like the room EQ app editor (Audyssey makes a $20 app for their system, Anthem Genesis allows you to adjust their curve, etc). Or do it in a miniDSP, even adding a BassEQ system, just don’t re-run your room EQ system with a miniDSP’s house curve in place, because the room EQ system will reverse it! Similarly, if you run something like Audyssey at this point, the system will try to change your crossovers and your sub “distance” (delay) setting, so you may need to go in and correct those.
Final thoughts.
There are a myriad variations to the method described here. But the concept remain very similar.
MiniDSP: A popular tool is a minidsp instead of using the phase adjustment on the subs to time align the subs. Some people also like to use the minidsp for EQ to take some of the effort off the room EQ system, or even use the minidsp to align the subs to the mains (distance tweak). And / or they use the minidsp to also apply bassEQ. I have a minidsp and I found I didn’t need to use it, and I wanted to share the process without using one.
MSO: Especially if you have three or more subs, including mismatched subs, and the room symmetry doesn’t lend itself to simply placing the subs in their proper locations for management of room modes, you can use a free piece of software called the Multi Sub Optimizer to take acoustic measurements of each sub, and figure out optimal settings to align and EQ your subs, that you then dump into your MiniDSP…..resulting in one great measuring single virtual sub that sounds good across multiple seats.
Tactile response: The gold standard is to generate enough SPL with your subwoofers to hit reference level in the LFE channel, 115db. If you have too large a room, too small a budget, or too little acoustic isolation (where that level of bass is going to disturb other people) you can consider using “tactile transducers” to get the feeling of extreme bass with fewer drawbacks.
Footnotes:
*In addition to my short summary at the start of this process, see also Welti's papers on this, his engineering papers in which he discusses tradeoffs between output and consistency across multiple seats, his interviews like at AVRANT on YouTube, etc.
**Don't fall prey to magical thinking about stereo subs, speaker level inputs, and so on. You don't need to blindly follow any theories, either. Rather, make use of the double blind studies conducted with hundreds of people, that show how a mono subwoofer signal, processed correctly, coming from the right places in the room, is seamless and transparent to the source.....even more than full range speakers, stereo subs, and other kinds of superstition left over from the 20th century, which have no support once listening tests are no longer sighted.
An excellent series of short articles by Anthony Grimani can be found here: https://grimanisystems.com/articles/ with direct links:
part-1
part-2
part-3
part-4
part-5
*** I realize that for some people, a single sub is the only option. If this is due to budget, frankly, I'd split the budget and get two lesser subs. But for the sake of completeness, I'll mention that you can increase your chances of decent sound from a single sub with a few simple steps, if you only care about one seat, where a single subwoofer can work okay sometimes. The following is an abbreviated process and assumes you don't want to buy a calibrated microphone and learn to use REW for measurements.
1. Place the sub on your main seat.
2. Then play some “bass sweeps” from youtube orCustom Chirp and Sweep Tones | Audio Test File Generatoron repeat. 10hz to 200hz is enough.
3. Go to the places in the room where you might place the sub. Using your ears or a free SPL meter app on your phone, see what the sweep sounds like from that location.
4. When one of those places gives you a nice smooth bass sweep, that is a good place for the sub. (What does a smooth sweep sound like? Use your ears or an SPL app on you smart phone. You want to see the SPL dip as little as possible. You probably won’t find some place that is perfect. So settle on a place that has the fewest big long deep dips. Even your ears can be enough for this.)
5. That place with the fewest long deep wide dips is where to place the sub.
6. Then run your room correction, and enjoy.
***See the discussion of RP22 for more about output.