[Letitroll98]

Note that you already have three subs, these two and your 10", and there is no requirement for four subs in a multi sub layout, only that four is easier to integrate.  Three can work just fine.  Something you already have is the most Cheap and Cheerful of all. 

[jtwrace]

four subs in a multi sub layout, only that four is easier to integrate.  Three can work just fine.

Not really true.  It's once again all room dependent.

[Rclark]

They will work. Qts of 0.47 isn't that low, and you'll need power and EQ. But you are going to need those anyway.

 You think so John? Would you be willing to help me on that? I would LOVE to see those beasts in a big H frame or on some baffles. That would be epic.

 And to JTTrace.. Wow!

 So you have four subs, two pairs of different types, huh? I can only imagine how that sounds. You've had OB's before right? It seems like the upper echelon uses either OB or multiple subs, but it seems like OB is falling out of favor for the multi sealed sub approach.

[Russell Dawkins]

But then, who has tried a swarm of OB subs?

I want to try a pair of OB subs placed close to my main speakers and a pair of other subs arranged asymmetrically, both IB (in two different walls and heights).

I expect to get 35 clean Hz extension from the OB dipoles and approx 10Hz from the IB monopoles.

[mcgsxr]

QTS of 0.47 will work on OB for sure - are they single or dual voice coil?

[jtwrace]

And to JTTrace.. Wow!

 So you have four subs, two pairs of different types, huh? I can only imagine how that sounds.

Awesome!

You've had OB's before right?

NO!

It seems like the upper echelon uses either OB or multiple subs, but it seems like OB is falling out of favor for the multi sealed sub approach.

That's correct.  OB is NOT the way to go for multiple sub system.  Remember, it takes about double the about of drivers for the same output.  To me, that's one of the disadvantages.

AC member TomS had OB's and has switched.  Maybe he will chime in here.

One last thing, if one is not willing to measure then they're wasting their time with the multi sub system.  You can't do it by ear and get the best result.  I think you should measure no matter what but when you have a multi sub thing going on there is A LOT of stuff that can be tuned for the best sound.

But then, who has tried a swarm of OB subs?

See above.

[JDUBS]

That's correct.  OB is NOT the way to go for multiple sub system.  Remember, it takes about double the about of drivers for the same output.  To me, that's one of the disadvantages.

Do you believe that the only disadvantage is that it takes double the amount of drivers for the same output?  If so, its not a big deal.  H-frames aren't all that big.

-Jim

[TomS]

Pre-warning: My setup is not really cheap and cheerful, but many of the same ideas can be applied  Also, I don't have the expertise of Duke, Dr. Geddes, Danny, etc., just a curious user. Duke would probably add that "swarm" is just one specific implementation of multi-source bass and there are many other means to the end - great bass!

I am very enthusiastic about the benefits of having a multiple bass sources in the room, but it does not mean that you can just throw anything in there wherever you want and expect great results. It takes a lot of work and discipline. There are different schools of thought on how to approach this, but there are some common themes. While the output requirements of each sub are less, you still must use decent "quality" subs - drivers, rigid braced enclosures, sufficient clean power, etc. Unfortunately, that may fly a little in the face of "cheap and cheerful". 3 is good, 4 or more is even better if you have the room, but it can be very challenging to tune. You MUST measure and be patient as cause and effect is not always obvious. Over time, it starts to make sense, but it was a LOT of time for me.

I currently have been experimenting with:
- 2 GR OB dual servo subs (Rythmik plate amps)
- 3 Geddes bandpass subs with passive radiators (Crown K2's)
- 1 GR sealed servo Dymaxion sub (Rythmik plate amp)
- 1 SVS CS Ultra ported sonotube sub (Samson S1000)
- 1 sealed dual 18" ULF hopefully coming soon (Crown K2)

All are driven summed L+R mono via a stock DCX2496 lowpass or bandpass crossover setup (6 sub outputs, individually tuneable). In the next couple months I hope to be doing a lot of trials and measuring to continue my own learning experience of how this works and how each responds to tuning.

The OB bass sources can certainly be used in this setup, however one needs to realize they are rather limited in placement, generally to the plane across the front, out from the front wall 3-6', or at the back plane (though that is often where the listening position is) depending on your room shape. There are nulls to the left and right sides of the OB's. For my narrow room setup (13') this is right where the Geddes Abbeys sit about 5' out from the front wall and directly adjacent, along the side walls, beneath or between them. One problem with this is that it is all about a multitude of bass sources spaced apart. The Abbeys (sealed 12" monopoles) are already providing upper bass from the very same spots. Note that the mains count as 2 bass sources also, since we're talking about up to 250hz or so! Sooooo, the options for those OB's just became rather limited. Once I returned to monopoles placed at specific, though seemingly random, spots around the room, things started to improve. I also have different XO points for each one. I suspect the resulting very high model density is what makes the bass sound very natural and seem to have limitless headroom. I doubt any of the individual subs are working very hard but they can if they need to. Huge X-max and 2kw amps aren't really needed unless maybe it's the only bass source in a certain ULF band. In my present setup, even without 18" ULF, there is plenty of visceral impact when called for.

If the OB bass system at the mains location is producing all of the primary bass, such as an integrated system like GR Super V's, V1's, V2's, Hawthorne Duo/Trio, Emerald Physics CSx.x, Orion, et al, I suspect the mixed approach might work better. I believe Danny did this with supplemental ULF sealed servo subs (<25hz?) with the Super V's at RMAF this year, which I would imagine works pretty well. As I recall, the Orion/Thor is another example of this approach.

Basically if the OB bass works at those primary locations, you are in luck. Otherwise they aren't going to give you as much placement flexibility when used as supplemental bass/subs, which sort of defeats one of the tenets of multiple bass sources with random placements. In a very wide room setup they could be spaced out further, but it just isn't possible in mine.

Oh, also realize that this applies not only to lateral placement, but vertical positioning too. For anyone with 8' ceilings, you're going to see problems in the 72-76hz region. Just check your mode calculator or do a quick measure to see the impact in your own room.  Placing a sub midway or 2/3 up the wall introduces yet another element that is very difficult to implement in the real world (they're heavy!). It is yet another tool at your disposal though if you want to try it.

In terms of the total mix of subs, my bass sources end up being a fairly constant distance from each other so they are evenly spaced around the room, though not at all symmetric. This yielded smoothest response without EQ. Anything nasty below about 40 hz can be dealt with global PEQ as long as you're not boosting too much which takes power quickly. Above that it's different at various listening/measuring positions.

It all starts with an open mind and an inexpensive measurement system. I'd say give it a try, even with cheap and cheerful subs (OB's too) to see what happens.

So many variables, but so worth it ...

[Rclark]

TomS I think your style of setup is what I aspire to.

Anyway, after reading all that, it appears that "random placement with differing elevations" didn't work in your room?

 I got the impression the swarm was easy to set up, that you placed the 4 subs randomly and that it didn't really matter so much where they were placed. I think someone said they tried arranging the subs in different spots and it all pretty much sounded the same.

 I think meanwhile I'm going to bug JohnR about getting these Brahma's in an H frame. If that's possible, then I think I will try a swarm later on as I learn. I might actually just do a swarm next in the living room system with my monitors. In fact, if I can get these Brahma's as OB bass in my room, then that's what I'll do. I'm giddy at the idea of seeing these bad boys in action.

[TomS]

TomS I think your style of setup is what I aspire to.

Anyway, after reading all that, it appears that "random placement with differing elevations" didn't work in your room?

 I got the impression the swarm was easy to set up, that you placed the 4 subs randomly and that it didn't really matter so much where they were placed. I think someone said they tried arranging the subs in different spots and it all pretty much sounded the same.
...

Actually no, quite the opposite and sorry I wasn't clear.

Elevating the sub is another possibility if "conventional" placements don't quite get it done.

It DOES matter a whole lot where they are placed. That is the first method to achieve a relatively flat response.

Also, it may have "sounded the same" to someone, I seriously doubt it measured the same. You would have to get very lucky to drop down 4 subs and get a smooth result by chance. Until you've really done it right  you don't know how good it can get. I know because I've been through all that several times until I was exhausted.

With all that said, Duke's Swarm setup may be entirely different in that regard.

[Rclark]

 Well, this guy never emailed me back, and you guys have me back at my senses. The ability to make a thread is a great repellent for the impulse buy.

 I need to go in on monday and possibly buy a new charger. If I can I'll try and swing by rat shack for an spl meter, and hit up Fry's for a cheap external soundcard that can mate up with it (I run a cheap laptop, nothing more than headphone/microphone minijack connectors), dl some software, and start measuring. When I finally have something, I think I will post it here.

[TomS]

Great idea! I am learning as I go but have a lot of help here. I might post on the acoustics circle for more inputs on your room, etc. You'll get a LOT more traffic on it.

[jtwrace]

I got the impression the swarm was easy to set up

As I and Tom have noted, that's not the case.  Is it impossible?  No.  As I've noted in many places of this forum if one is not willing to measure they're really just wasting their time.  I think after seeing what I was able to accomplish it gave Tom some motivation to chase the line.  In the end, he got it really nice and is very happy. 

[Rclark]

Oh so you're the instigator. Well, with a graph like that....

[JohnR]

I need to go in on monday and possibly buy a new charger. If I can I'll try and swing by rat shack for an spl meter, and hit up Fry's for a cheap external soundcard that can mate up with it (I run a cheap laptop, nothing more than headphone/microphone minijack connectors), dl some software, and start measuring.

While SPL meter is not a bad thing to have, for this I think you'll need a mic like the Behringer or Dayton from Cross-spectrum labs. Read all about it here: http://redspade-audio.blogspot.com/2010/08/basic-measurement-setup.html

You will also need a mic preamp that supplies phantom power. Paul (above link) prefers a separate mic preamp as shown used with the computer soundcard; I prefer an external USB preamp/soundcard. I use a Roland UA-25EX, but there are cheaper options, while it's getting off-topic perhaps others could chime in with the hardware they use.

[jtwrace]

While SPL meter is not a bad thing to have, for this I think you'll need a mic like the Behringer or Dayton from Cross-spectrum labs. Read all about it here: http://redspade-audio.blogspot.com/2010/08/basic-measurement-setup.html

Agreed.  I use the ECM from Cross too.

while it's getting off-topic perhaps others could chime in with the hardware they use.

I do agree about the OT but like you I find this so important.  There is more to gain with meassurements then swapping out your hardware.  In fact, once you mreasure you will get more return if you do swap out the hardware. 

I use the M-Audio MobilePre which seems to work well.  HolmImpulse is my choice of software and like REW it's free.

[JohnR]

I think meanwhile I'm going to bug JohnR about getting these Brahma's in an H frame.

You don't need an H frame to get started, just put a driver in a baffle of say 18" square, and screw a couple of bits of wood on the side to hold it up. Then use any amp you have available and start measuring in different room positions. Once you have some data you can decide on whether to build a solid H-frame, or go for boxes.

[JohnR]

There is more to gain with meassurements then swapping out your hardware.

Hi, I meant measurement hardware With these things it's good to know what is working for people. You answered anyway 

[JohnR]

Pre-warning: My setup is not really cheap and cheerful, but many of the same ideas can be applied 

Thanks for an excellent report. One comment:

I suspect the resulting very high model density is what makes the bass sound ...

Modal density is a property of the room, the number or position of subs doesn't change where the modes are.

[Duke]

I got the impression the swarm was easy to set up...

I'd like to differentiate between my commercial four-sub system, which I call the "Swarm", and a fullbore Geddes system; the former is simpler and the latter more advanced. 

My system uses a single central amplifier to drive four small subs.  The four small subs have an inherent rolloff that is approximately the inverse of anticipated room gain.  The single central amplifier has a 4th order lowpass filter to aggressively roll off the top end of the subs.   I suggest a few rules of thumb for placement, fiddling with levels and crossover frequency and (if necessary) using the single band of parametric EQ in my amplifier.

If I understand correctly, Earl uses three or four subs which each have their own amplifier, and he takes advantage of that to optimize each sub individually for level, low-pass frequency, and phase.  He is also using mains that are very low-Q sealed box systems, so he uses a different protocol for getting a good blend with his mains.