Something I compiled a while ago which I feel is relevant to this thread:
It is basically a rehash of several experts (Earl Geddes, Todd Welti, Devantier, Floyd O'Toole) in the industry, (and one long term enthusiast and colleague - Duke LeJeune) whom I have credited at the end of this rather long soiree.
None of this material was originally authored by me.
Regarding the setup of multiple subs. The minimum required is 3, the maximum most people use is 4, although you could use more. Once you use 4 or more, placement starts becoming far less critical.
Let's do some background first, and then come back to talk about bass in small rooms.
Background TheoryThe first thing to know is, the ear/brain system can "average out" peaks and dips that are fairly close together (like within 1/3 of an octave of one another), but cannot average them out if they are far apart. Second, the ear has poor resolution in the time domain at low frequencies, such that it takes at least one full cycle to detect, and several cycles to hear the pitch of, bass energy.
Now with this background, let's look at bass reproduction in a small room (our home listening rooms are "small" in comparison with an auditorium). Starting with a single subwoofer, we measure it in-room and find a disturbing peak-and-dip pattern; plus 6 dB and minus 9 dB swings are not uncommon. These are totally room-induced, and we can change and re-arrange them by moving the subwoofer or the listener (microphone) position, but we cannot make them go away.
Why are these bass region peaks and dips so bad? Don't room reflections also cause nasty peaks and dips at midrange and treble frequencies? And since the peaks and dips aren't present in the first-arrival sound, don't they look worse on paper than they really are?
While it's true that unsmoothed in-room measurements up into the midrange and treble region look like hash because of reflections, perceptually it's not so bad because 1) the ear detects the pitch quickly enough that these reflections have minimal effect on pitch definition; and 2) at these frequencies the room-induced peaks and dips are close enough together that the ear averages them out. Unfortunately in the bass region, it takes several cycles for the ear to even begin to determine the pitch and by that time the room's effects are in full cry. So perceptually we cannot separate the subwoofer from the room; they form a system. Second, because the room is physically fairly small relative to bass wavelengths, the reflection-induced peaks and dips are far apart - too far apart for the ear to average them out. So the rest of the spectrum may sound great, but the bass sounds artificial.
Equalization seems like a logical solution, and indeed can produce significant improvement at a single listening position. But in most cases the specific bass problems are local, and equalization affects globally (throughout the room), so fixing the problems in one location will usually make the problems in another location worse.
The solution is to use multiple low-frequency sources scattered asymmetrically throughout the room. This way each will produce a unique peak-and-dip pattern at the listening position, and the sum of these dissimilar peak-and-dip patterns will be significantly smoother than any one of them alone would have been. The remaining peaks and dips will not only be smaller, they will also be closer together so the ear/brain system has a better chance of averaging them out.
Just to be clear, a distributed multisub system not only reduces the variance at a given listening position, it also reduces the variance between different listening positions. This makes it easier to equalize the in-room response because any remaining significant in-room problems tend to be global instead of local. The drawback of course is the initial cost of the multisub system, and the real estate it occupies.
Now let's look at a big room for a moment - like an auditorium, or a good recital hall. In this case the room boundaries are far enough away that the in-room peak-and-dip pattern in the bass region looks more like our small room would in the midrange region; that is, the peaks and dips are much closer together, so the ear can do its averaging-out trick. This is one of the reasons bass often sounds much more natural in a big room.
As a general principle, frequency response at the listening position is the dominant factor in perceived bass quality. Transient response is audible to the extent that it affects frequency response, but not much beyond that for a reason that we've already seen: The ear has very poor resolution in the time domain at low frequencies.
So at what low frequency does the ear start to have poor resolution in the time domain? This is called the Schroeder frequency or transitional frequency.A great majority of small rooms (i.e. less than the size of an auditorium), has the Schroeder frequency at around 150-200 Hz. Because we can't localize sound sources below 150 Hz the use of additional subwoofers can smooth the frequency response at a given listening position or over a certain listening area.. Remember, a great, great, great majority of recordings do not have bass recorded in stereo. When you perceive that the bass is coming right of center or some other point in space, I think its because another higher frequency was being played at the same time and that is what you are localizing to. That's why I chuckle at the notion of 'stereo bass.' I ask, what frequency are you really talking about? What is a bass frequency? It's not 500 Hz mind you. It's much, much lower. For example if a bass guitar player, plays the lowest "E" string, it's about 42Hz. I cannot perceive any localization at that frequency. However, if he were to play that frequency along with one that is a few octaves higher, I may be able to appreciate 'localization.' I am not totally comfortable with this, but I believe the Haas precedence effect, and the Summing Localization effects have some input to this discussion.
One final thing that matters is the effects of room gain. Briefly, the room will usually boost the low frequencies due to boundary reinforcement; as we go down in frequency more and more room surfaces fall within 1/4 wavelength of the source and thus we get more and more in-phase reinforcement from those nearby reflections. The commonly perceived superiority of sealed over vented boxes is because the more gradual rolloff of a sealed box is a better matchup with typical room gain. Outdoors where there is no room gain, vented boxes sound better - anecdotal evidence that it's the frequency response at the listening position, rather than the transient response (an area where sealed boxes excel), that has the biggest effect on perceived bass quality.
ApplicationOkay, now that you've got a background, lets talk about what Earl Geddes recommends for ideal bass. This is different from other experts such as Todd Welti and Floyd Toole of Harmon Kardon. Their research and algorithm for minimizing spatial variation of the low frequency response by using symmetrical bass sources over a broad designated area is described in great detail in the 1st article listed below. Earl recommends asymmetrical placement. The advantage or difference? Only three subs are needed and you don't need to place the subwoofers at specified locations. Nonetheless there are some basic rules that have proven to yield ideal results:
Put one sub in a corner close to the mains. The second sub is a lot more flexible as to its location, but it should not be in a corner. Side wall or back wall, near the midpoint is a good idea. Put the third sub wherever you can that is not too close to the other two. It's a good idea to get one of them off of the floor. If you want to get real stingy, Earl recommends getting that sub more than 1/2 way between the floor and ceiling, although this isn't possible in all locations.
If you want to go forward with this you'll need to purchase a reasonable accurate and calibrated microphone (like a Cross Spectrum labs Behringer ECM-8000), micpreamp, boom mic stand, etc...and have a thorough read of the 2nd article, which goes through the calibration and setup procedure in detail.
Hope that helps
Anand.
P.S. I should give credit where its due...without the tutelage of Earl Geddes, Duke LeJeune, Todd Welti or Floyd O'Toole, none of this would've made any sense to me...
ArticlesTodd Welti/Floyd Toole's/Devantier's Symmetric subwoofer placementMarkus Mehlau's calibration procedure for triple asymmetrical subwoofer placement per Earl GeddesA more simply worded article on the use of symmetrical subwoofer placement by Todd Welti
