|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
[ Asylum Support ] [ Rules ] |
Model: | The Swarm |
Category: | Speakers |
Suggested Retail Price: | n/a |
Description: | 4 speaker subwoofer array with amp |
Manufacturer URL: | Not Available |
Review by Mickey Bob on April 02, 2007 at 19:21:33 IP Address: 204.128.192.3 | Add Your Review for the The Swarm |
I recently had the opportunity to evaluate Duke LeJeune’s Swarm Subwoofer System. I use the word ‘system’ advisedly as the Swarm consists of five components: a mono plate amplifier and four 5” subwoofer modules. The Swarm is essentially a surround-sound subwoofer system.
The design principle behind the Swarm is simple. Four 5” drivers have the same area as one 12” drivers but.. and this is the whole point … four small drivers are much easier to drive than one large one. Something similar is done in the area of astronomical optics where increasingly a number of smaller mirrors with separate aiming drives are replacing the giant single mirrors used in telescopes of the past.
Just as it’s easier for smaller mirrors to be operated to exacting standards, it’s technically easier for small loudspeakers to reproduce sound accurately. One of the challenges in subwoofer design is the mass represented by a large driver. Once in motion it wants to stay in motion and yet we expect it to switch directions - in and out - 60 times a second for a 30 hertz signal. The more massive the driver material, the less accurately the driver can trace a delicate waveform. Another challenge to the subwoofer designer is the rigidity of the driver material. The more a driver cone flexes, the more it will smudge sonic details. Some subwoofer designers use exotic carbon fiber designs to get a combination of virtues - lightness and stiffness - but the cost of these materials is high.
All these technical challenges are easier to meet with a multiplicity of small drivers each of which is less massive and less prone to flexing. And yet, when the sound out put is aggregated the result is equivalent to a much larger driver. In the case of the Swarm, the four drivers are roughly equivalent to the surface area of a 12 inch subwoofer.
Thus does Duke LeJeune address the technical issues associated with constructing a quality subwoofer. But there is an additional element to consider - room acoustics.
Anyone who’s ever tried to find the best position for a subwoofer knows how difficult it can be. The subwoofer output interacts with the room acoustics in unpredictable ways with the result that different locations in the room get varying degrees of bass. At some points, out of phase reflections meet and cancel each other out. At other locations, compression waves collide and artificially boost the amplitude of certain frequencies. And then you have also to contend with room nodes - those frequencies at which the structural elements of the room resonate adding another befuddling variable to the sonic environment.
By dint of trial and error you can ultimately find a position for the subwoofer which will yield optimal sound at your selected listening position. But the sound quality will change quickly as you move away from that position. With one subwoofer, you can get it optimal in one place at the cost of getting it wrong everywhere else.
LeJeune’s Swarm system seeks to address this problem of room integration through its multiple driver design. The Swarm subwoofer array is intended to be installed asymmetrically. LeJeune reasons that the multiple drivers will create multiple resonances, multiple cancellations, and multiple compression wave collisions. But not only will these various effects be multiple - they will be random due to the asymmetric placement of the speakers. LeJeune hypothesizes that the effect of these multiple, random room interactions will be a soundfield where the various reflections, resonances, and distortions tend to cancel each other out resulting in a consistent sound throughout the listening environment.
The Swarm’s unconventional design includes some new thinking in the crossover design as well. LeJeune has chosen a fourth-order crossover. The salient feature of a fourth order design is a much quicker rise in output compared to other designs. The goal here is to keep the sub’s output below that of the full-range drivers and avoid the ‘smearing’ of sonic details often heard when a sub’s output intrudes on the frequency range of the main speakers.
So… how does this all work out in practice? I would say, based on my experience… pretty well. When paired with my Quad ESL-63’s running full range the Swarm added a nice layer of bass reinforcement right where the Quads needed it. Most importantly, by adjusting the volume and the crossover point, I was able to bring the bass in at just the point where the Quads seemed to roll off. I can’t overstress the importance of this. In other experiments with my Quads, subs always seemed to produce too much output in the mid-bass region where the Quads actually do quite well. When a sub adds its output to this region, it degrades rather then enhances the Quad sound. But with the Swarm, I seemed able to dial the bass in just beneath mid-bass region giving the overall sound more depth and a very satisfying amount of bass impact without compromising the Quads’ legendary transparency and clarity. (I had the crossover set to what I estimated to be the 60 Hz position, and the volume at just below 25 percent.)
The Quads have always excelled at reproducing acoustic music, but with the Swarm added in, the Quads could really rock! The kick drum now really had a kick. And bass guitar had just the right amount of added heft I’ve missed from the Quads in the past. And once again I noticed that none of what the main drivers did well was being compromised by the Swarm. I could keep its output below the main drivers so I could still enjoy their high resolution with the added benefit of a real kick in the bottom. Steely Dan and Roy Orbison really came alive with this configuration.
After listening to some of my favorite CD’s, I decided play around a little bit with a home theater set up. Out went the Quads and in came the Sound Dynamics RTS-3’s. (This is the ‘affordable’ monitor Harry Pearson raved about a few years back.) With this more dynamic speaker I could really crank up the volume. Of course it doesn’t play as low as the Quads so I had to dial the crossover point up to about 90 Hz, and raise the output dial to the high noon position.
Wow. Now admittedly, this was just a stereo system, but with the Swarm helping out, the Sound Dynamics speakers were doing something they’d never done before - throw up large stable images with a vertical as well as horizontal dimension. Symphonic music took on its proper scale. Now don’t get me wrong. In every way the Quad/Swarm system was the overall best for music. But combining the Swarm with the Sound Dynamics gave me a system on which the sound assumed the large scale proportions you need for dramatic home theater. I was able to experience prodigious amounts of appropriate bass quality and quantity. In ”Superman Returns” the planet Krypton exploded not just with a one-note bassy blast but with a palette of bass colors, some actually inaudible. In “Hunt for Red October” (a film which boasts some of best recorded and engineered sound I’m aware of) the Swarm reproduce the various engine noises and ambient environments of the ship settings with a vividness and impact that really enhance the dramatic experience.
As you can see, when properly integrated, the Swarm do an awful lot right without doing much wrong as far as I can hear. Still, there are a few things on the negative side of the ledger. The system would benefit from a remote so you could experiment with various crossover and level settings from your listening position. However… the exercise of getting up and down probably was good for me and I must say that eventually in both cases (Quads and Sound Dynamics) I arrived at what I considered optimal settings and was no longer tempted to keep tweaking. The other point really has nothing to do with the Swarm but everything to do with room acoustics. No matter how carefully you adjust and tweak there will come, from time to time, bass notes that will resonate unpleasantly loudly in your room. The fact is your walls, or ceilings, or room contests are going to be susceptible to various resonances that will cause certain bass frequencies to jump out too loudly. The fix for this is probably some kind of equalization - a feature that some subs do offer. But like the convenience of a remote - this is something that would greatly increase the cost of the system.
On balance, and in summation, the bottom line on my experience with the Swarm is…. If you can spare the room for a four speaker sub array, then the Swarm - once it’s in production - will be an attractive (and, I suspect, a relatively affordable) option. If room is tight, perhaps Duke can be persuaded to also produce a three speaker version, although I understand that the reason he chose four speakers is that it’s the point of ‘optimal’ return. Fewer makes a big difference in quality, more doesn’t.
Product Weakness: | Four speaker units to contend with, no remote operation |
Product Strengths: | Prodigious but musical bass, steep crossover curve that helps it nicely integrate with Quad speakers, relatively immune to placement |
Associated Equipment for this Review: | |
Amplifier: | Creek 5350 |
Preamplifier (or None if Integrated): | none |
Sources (CDP/Turntable): | Sony mass market cd player |
Speakers: | Quad esl 63; Sound Dynamics RTS 3 |
Cables/Interconnects: | audioquest, straightwire, monster |
Music Used (Genre/Selections): | classical, steely dan, home theatre |
Room Size (LxWxH): | 12 x 12 x 8 |
Room Comments/Treatments: | carpeted floors, absorbent panels |
Time Period/Length of Audition: | 4 weeks |
Type of Audition/Review: | Home Audition |
Follow Ups:
Looking into things a bit further, the woofer is 6.5" diameter. Four of these would add up approximately a 12" driver. Rulers can sometimes come in handy, as this was not a subtle difference.One of Earl Geddes' relevant papers can be found at http://www.gedlee.com/downloads/sub study .pdf
Some of the underlying assumptions may be flawed, however. Incidentally, you should try to have one of the drivers near the ceiling. Good luck with that.
Again, the dimensions of your room are INTENSELY problematic.
Placing one subwoofer near the ceiling is optimal but not essential to reap benefits from the scattered multiple subwoofer configuration.In my little subwoofer modules, the driver and port are both on one of the 20" by 9" enclosure sides, one at the top and one at the bottom, with the terminals in between. I recommend placing two of the enclosures woofer-on-top, and two port-on-top. This gives a little bit of staggering in the vertical plane - not as much as placing one module near the ceiling, but still better than none at all.
Although I didn't do this in the set Mickey Bob had, the port tuning frequencies will be staggered a little in the production version.
Hmmm. Will trying posting that link again
Ah, didn't see Duke's contribution until after posting mine. That clarifies the driver question.
The Swarm subs are based on work by Earl Geddes. If I recall correctly, however, at least one should be installed off the floor in order to address the floor-ceiling modes.Also, don't four 5" drivers add up to the area of a single 10" driver? Pi-r-squared and all.
Looking at your room dimensions, no wonder you have trouble with resonances!
You are absolutely correct, the scattered multiple subwoofer configuration is based pretty much entirely on conversations with Earl Geddes. In fact, Earl has applied for a patent on a subwoofer system whose principles I'm using - and just for the record I built the Swarm with Earl's knowledge and with permission to incorporate his concepts. Several other researchers have recommended multiple subwoofers, though usually with more symmetrical placement (Welti, Griesinger).The woofers are nominally "6 inch" units. The frame diameter is nearly 7 inches, and the cone area is 140 square centimeters. Four of them would be 560 square centimeters, which is about right for a single 12" woofer.
The woofers I use have a theoretical maximum linear excursion of 13 millimeters. With the vented alignments I use, they don't reach 13 millimeters excursion anywhere within their passbands, and the plate amp has a high pass filter that should protect against over-excursion below the tuning frequency.
A 5" driver is really has 4" cone, a 10" driver has an 8" cone, a 12" driver is really a 10" cone.2 squared x 4 = 16, 4 squared = 16, so 4x5" = 1x10".
It's got to be bigger drivers than that. Who sells a 5" subwoofer? You'd need at least 4 8" drivers to be worth much, but 10" or 12" makes more sense because you're going to get a lot of cancellation from the "swarm" approach. You could do a "swarm" with NHT's modular U2 setup with 4x 12" for $2500.
The ideal setup would be 8 woofers, each in a corner, floor *and* ceiling. That would be cool.
I'd have preferred to use larger drivers, all else being equal - but all else was not equal. One of my concerns was price point, and another was enclosure size. The 4 X 12" NHT system would probably kick my system's little butts (all four of them) in many areas, but mine has sufficient extension for most music and sufficient headroom for use with most planar systems (Maggies and Quads).Of course I have a "SuperSwarm" on the drawing bored, but want to see how this one does in the marketplace first.
Just for the record I cannot agree with your suggestion to place subwoofers in corners if the goal is smooth bass. Corner placement will give you loudest-deepest-possible bass, but also maximize excitation of room standing wave modes so you'd have relatively pronounced (and therefore audible) peaks and dips. The more randomized location approach I advocate results in smoother bass because each sub will generate a unique peak-and-dip pattern at any given listening position.
Below is a link to a simulation run by Earl Geddes comparing corner placement of four subwoofers with randomized placement of four subwoofers. Each of the different-colored curves is the summed bass response for a different listening position.
Duke
Well, you can't really translate 4 subs in corners to subs in all 8 corners. In theory, that would largely cancel all three major modes. In theory, I've never tried it, it would be a hassle hanging 4 of the subs. Besides, I've never been a fan of subs in the corners anyway and prefer the more "random" method in practice. I just haven't tried my 8 corner theory yet.
Intersting that you have gravitated towards a more random configuration. I didn't exactly gravitate; more like someone hit me upside the head with ye olde two-by-four and then it made sense to me.How would having subs in all 8 corners cancel all three major modes? I don't see how it would.
Thanks,
Okay, let's say you have a 20' room that has a 28Hz (40') standing wave. The bass is coming from the front, it bounces off the back wall. As it bounces, it's at the rarefaction stage. At the same time, the subwoofers are all beginning to reproduce the compression state of the 28Hz wave. So, the sub in the back is putting out a 28Hz compression that is lining up with a smaller 28Hz rarefaction and it cancels it, leaving only a little 28Hz energy. This works for lateral and vertical waves because there is a "cancellation" device at every room boundary. Of course, there are other ways to do it and this may not be the best since I haven't tried it, but just a theory I have. Not quite the "active cancellation" I have, but I'd be curious if someone has ever tried this.
I don't think that would work.I think you would get some smoothing, but I think the two bass pressure waves will largely pass through one another rather than the one neatly cancelling out the other throughout the room. The simulations and measurements I have seen for subwoofers in corners doesn't show evidence of global cancellation of the primary room modes. Recall the link above showing subs in four corners - it looks to me like more than one room mode is still significantly excited, and if having subs in opposite corners suppressed room modes we'd expect that four subs in floor corners would only leave the floor-to-ceiling 71 Hz mode unsuppressed.
Asymmetrical placement of multiple subwoofers doesn't rely on cancellation - it relies on summation smoothing due to each subwoofer generating a unique peak-and-dip pattern at any given listening position.
You need 9.
Don't forget to hang one from the ceiling fan so as it spins,it can take care of "random" room nodes caused by rarefied, gaseous air caused by farting guests ??
.
> Looking at your room dimensions, no wonder you have trouble with resonances! <
Amen to that, brother!
I'm not sure how it adds up - I may have the driver diameter wrong. FOr specifics on the dimensions and any technial matters, give a holler out to Duke, a frequent participant in this forum.For most of my evaluation I had two of the speakers off the floor - about two feet off.
Dukes a very helpful guy.Of late he has had some very interesting ideas on audio reproduction. Dont see how you could go wrong purchasing from him.I do perfer the low excursion bass system designs but these are costly and tend to be very large think bass horn;) Duke swarm idea is a good one.
Thanks for a well written review that any audiophile can relate to.
I have owned the Quads, and used the excellent, but expensive Entec Subwoofers with them.
I know how hard it is to get a seamless blend.
That is one reason I am partial to pre amps with a remote subwoofer volume control!
I wouldn't be w/o it ever again.Audio Relatives and Neices or whatever they are called must be using long X Max drivers to acheive what you are hearing.
Now, the magic question, How Much do it cost ??
...fifteen hundred dollars for the set. The price may go up over time if (when) my costs go up, but that's where it will start.The production version will have two plate amps instead of just one, so the bass will be in stereo. Now admittedly most recordings are mono below 80 Hz or so, but in cases where a fairly high crossover frequency must be used stereo has advantages. Also, the box-count will be down from five to four, as I'll incorporate the two plate amps into two of the subwoofer enclosures.
My Creek 5350 doesn't have any kind of volume regulated line output, so I had to drive the plate amp from one of my speaker terminals. I don't know if the results would have been much different with a line level out.As to the type of drivers - I don't know if they would be described as "long excursion" drivers. We'll have to ask Duke LeJeune, the designer, about that.
Duke?
As to price - Duke has mentioned a figure but I hesitate to repeat it in case he needs to raise it. He's still researching suppliers and deciding on whether to go with mono or stereo amps, but I'd bet that compared to the big boys, it'll be a bargain. For one thing, I think Duke is planning on selling direct to buyers rather than through a dealer network. He wants the cusotmer's money to end up in their systems and not get soaked up by layers of distribution.
Thanks, Bob.The plate amp in the Swarm has both line-level and speaker-level inputs. The gain may be different, but that should be the only noticeable difference.
The drivers used are "6-inch" nominal diameter, and have a theoretical maximum linear excursion of 13 millimeters. With the tuning I use and with the protective high pass filter built into the plate amps, their maximum excursion requirements should remain below 13 mm in this application.
I bet it will be fairly priced too ?
Duke is well known, liked, and respected here at the Asylum.
His main speaker line seems to represent a good deal for someone wanting to go in that direction.Being a long time Sound Labs Dealer, he has a unique opportunity to compare what he makes to one of the best speakers around, period.
Thanks for the thumbs-up, ka7niq. But mostly I just talk a good game, that's all.
This post is made possible by the generous support of people like you and our sponsors: