[jtwrace]

What about two kinds?  Two OB and two sealed? 

Would this be Dr. Geddes approved or would he say it doesn't matter? 

Actually what is Dr. Geddes opinion on OB subs with his speakers?

[Tyson]

Peaks are resonances, and resonances are bad.  Having a lot of resonances is not, IMO, better than having fewer resonances. 

[poseidonsvoice]

What about two kinds?  Two OB and two sealed? 

Would this be Dr. Geddes approved or would he say it doesn't matter? 

Actually what is Dr. Geddes opinion on OB subs with his speakers?

He doesn't care about what low frequency alignment you use. What he cares about is value, and in his research, the dipole subs have rendered the lowest value when used in the *application* of multiple subwoofers in an asymmetric distribution in the room. However, for his own purposes he likes high sensitivity subs that are bandpass with 2 that have  broader frequency range and one that has a narrow and low frequency range. He stresses value almost all the time. But clearly the technique works for any low frequency alignment, be it dipole, bandpass, sealed, ported, etc...you are definitely not married to one alignment or the other.

Anand.

[JohnR]

Dipole subs would have the limitation that they can't be placed in corners, though... (right?)

[JohnR]

Having multiple subs excites MORE room nodes, but they are spaced at closer intervals, so don't sound as offensive.  But, a room mode is a room mode, and they are best avoided, where possible.

The additional subs work to cancel the mode excitement, not to excite more of them. It took me a while to realize this, but I'm fairly sure that is what is happening when you look at the measured examples.

[Tyson]

Really, that would be surprising, because it seems like the nulls of one sub would have to align with the peaks of a different sub for that to be true.  And, the other question I'd have is whether it would hold true for all positions?  This was the issue I had with the DEQX and the DCX2496 - I could optimize things for one or 2 locations, but the bass would be worse in other locations. 

[brj]

It might be helpful to look up "destructive interference".

As one example, flip down to the image under "Constructive and destructive interference" on the Wikipedia page:

http://en.wikipedia.org/wiki/Interference_%28wave_propagation%29

Obviously, this shows a simplified example, but it illustrates the behavior.

[JohnR]

Really, that would be surprising, because it seems like the nulls of one sub would have to align with the peaks of a different sub for that to be true.  And, the other question I'd have is whether it would hold true for all positions?  This was the issue I had with the DEQX and the DCX2496 - I could optimize things for one or 2 locations, but the bass would be worse in other locations.

It's not quite clear to me what you mean by "the nulls of one sub would have to align with the peaks of a different sub " - a sub doesn't have nulls and peaks, those are determined by the room. If two subs excite a specific mode at different phase angles, then that excitation may be reinforced or cancelled depending on what those phases are. Taken to an extreme, this is the basis of setups like the "double bass array" system.

Different locations in the room will be in different places with respect to the modes. However there is also the direct sound from the subs as well as the modes, which is the other way that multiple subs improve the response - the paper by Welti that was linked earlier is specifically optimizing to reduce spatial variation, in fact.

This is in theory, I have yet to measure my own room as I have to build two more subs first

[Letitroll98]

Really, that would be surprising, because it seems like the nulls of one sub would have to align with the peaks of a different sub for that to be true.  And, the other question I'd have is whether it would hold true for all positions?  This was the issue I had with the DEQX and the DCX2496 - I could optimize things for one or 2 locations, but the bass would be worse in other locations.

Please help me out on this one Tyson as I was following your logic until here, so maybe I'm the one missing it.  I thought you were saying that OB subs excite room modes less because of their lobed response, which I agree with.  By the nature of said response, optimization would occur in limited locations, an advantage if your trying to get the best response at your listening seat.  But then you're asking if multiple sub's averaged response holds true at all locations.  As I understand it both methods would be optimizing for a given listening location, but swarmed subs, if properly implemented, would give an even response over a wider area.  Please feel free to correct me if I've got what you're saying wrong. 

In any case, a person implementing either theory without a strong knowledge base is likely to mess things up royally.  That knowledge being what we are seeking here perhaps this link will help:

http://www.linkwitzlab.com/rooms.htm

[Tyson]

JohnR and brj,
Thanks for that, I guess you learn something everyday

I'm not really against multiple sub locations as much as I'm in favor of OB bass as a good solution to difficult rooms.  Also, from a subjective standpoint, I've heard very good implementations of box bass and very good implementations of OB bass, and my preference is for OB bass.  Throw in my acoustically difficult room and OB is clearly a better solution in my space than box bass, even with multiple subs.

[jtwrace]

Dipole subs would have the limitation that they can't be placed in corners, though... (right?)

I would think so.  Anyone?

[turkey]

Box subs radiate in an omnipolar fashion.  This means the bass is bouncing off all 6 surfaces of a room (roof, floor, front wall, side walls, rear walls).  Open Baffle bass radiates bass in a figure 8 pattern, so it eliminates the side wall re-inforcement and nulls.

I once looked into this, and it appeared to me that a driver in free air radiates in a figure 8 pattern when in a large room (as compared to the wavelengths being radiated), and also where the listener or measuring mic is at a large distance, again relative to the wavelengths (in other words, far field).

In a small room, and we all have small rooms in the bass frequencies, the effect was to make bass appear as if it was radiated omni-directionally.

I think that OB, vented, sealed, bandpass, horns, etc. can all work, but some are easier to work with than others.

Not a perfect solution, but certainly a good solution.  IMO, you want to excite room modes LESS, not more, in an ideal world.  Having multiple subs excites MORE room nodes, but they are spaced at closer intervals, so don't sound as offensive.  But, a room mode is a room mode, and they are best avoided, where possible.

The room modes are there no matter what you do, although you can treat the room to damp them.

It's best to spread the subs out so that the modes are excited more or less evenly, so that one doesn't stand out over the rest.

[ebag4]

I'm not really against multiple sub locations as much as I'm in favor of OB bass as a good solution to difficult rooms. 

+1

[Letitroll98]

I once looked into this, and it appeared to me that a driver in free air radiates in a figure 8 pattern when in a large room as compared to the wavelengths being radiated, and also where the listener or measuring mic is at a large distance, again relative to the wavelengths (in other words, far field).

In a small room, and we all have small rooms in the bass frequencies, the effect was to make bass appear as if it was radiated omni-directionally.

I'm sorry, I thought Tyson was talking about bass problems in the 200 Hz area which I assume would be above the Schroeder frequency in his room.  Of course below fs we are dealing with mostly the pressurization of the room and directivity is not the topic, in which your statement is correct.  If everyone wants to have a discussion on subwoofers per say, perhaps we should start a new thread, which could be very interesting.   

[TomS]

I'm sorry, I thought Tyson was talking about bass problems in the 200 Hz area which I assume would be above the Schroeder frequency in his room.  Of course below fs we are dealing with mostly the pressurization of the room and directivity is not the topic, in which your statement is correct.  If everyone wants to have a discussion on subwoofers per say, perhaps we should start a new thread, which could be very interesting.   

I agree a new thread with about the last 20-30 posts would be good.  That would allow both Earl and other experts on this to jump in comfortably on just the multi-sub topic.  There is a lot to learn.

[jtwrace]

I agree a new thread with about the last 20-30 posts would be good.  That would allow both Earl and other experts on this to jump in comfortably on just the multi-sub topic.  There is a lot to learn.

Yes, that would be great if we could move these sub posts to a dedicated "Multiple Subwoofer Setup" thread.   

There is so much to learn on that topic it's insane.

[Letitroll98]

Ok guys, here you go, have at it.

[jtwrace]

Ok guys, here you go, have at it.

 

[jtwrace]

Does anyone here have the new (3 port) GedLee bandpass subs? 

[John Casler]

Box subs radiate in an omnipolar fashion.  This means the bass is bouncing off all 6 surfaces of a room (roof, floor, front wall, side walls, rear walls).

Hey Tyson,

I could be wrong, but I think ALL subs at a low enough frequency radiate omnipolar.  The Dipolar radiation with nulls to the sides is more an effect of higher frequencies which become more "directional" (beamy) as the frequency rises.

Open Baffle bass radiates bass in a figure 8 pattern, so it eliminates the side wall re-inforcement and nulls.

I know that is the common perception, but I think it isn't accurate as frequencies drop.  Air is a fluid, and if you move that fluid at a low frequency with a large driver it will affect (and move) ALL the air around it, including the air to the sides (nulls form at higher frequencies)

Not a perfect solution, but certainly a good solution.  IMO, you want to excite room modes LESS, not more, in an ideal world.  Having multiple subs excites MORE room nodes, but they are spaced at closer intervals, so don't sound as offensive.  But, a room mode is a room mode, and they are best avoided, where possible.

As I understand it the use of multiple subs is to create multiple bass energies that emanate from different locations.  The importance of this, is that EACH location reacts to the room boundaries from THAT location, and EACH has the potential for serious MODES based on the timing from "that" location.

By placing the subs at different locations it places the listener in a field where they experience the combined sum of ALL the good and bad sonics of each sub.  This sum SMOOTHS the BAD interactions by combining them with the good, again over all the differing responses.
 
So you do not have the WILD swings of the worst modes from a SINGLE sub at a particular listening position, you have ALL the subs interactions (and each will be different to the sub location and the listeners location) summed, thereby equalizing out both the good and bad for a relative smoothness.

This is just my opinion based on MUCH playing around with up to 8 subs in my room, and having a dimestore education in physics and psychoacoustics.