2. tech-talk
  3. 147852-121242

A DIY Swarm-based Distributed Bass Array for SOTA Bass

State of the art bass turns out to be a lot more affordable - and achievable- than you think. It just requires a new approach.

The physical realities of low bass in small rooms is discussed by Earl Geddes in "Why Multiple Subs?" http://www.gedlee.com/Papers/multiple%20subs.pdf

Basically, a 20Hz wave is 56.5 feet long. So any bass that low is bound to bounce off a wall long before it even becomes a full wave. Then as it bounces back it runs into itself, either canceling or reinforcing itself, and this all happens even before one full wave can be produced. This is just physics, or acoustics if you prefer, and happens in every room and always regardless of speaker design. As Geddes has shown the shape of the room hardly even matters. At residential sizes they are all pretty much the same. So the conventional approach of one or two subs cannot possibly work. It never could. It was doomed by the physical and acoustical constraints of home-sized rooms.

With four however it becomes almost trivially easy. Not in the sense of being perfect. Its not magic. The quality and performance of the four speakers still matters. But its trivially easy in the sense that the improvement inherent in using four is so great as to swamp these details.

Or at least that is the conclusion that led me to build my own distributed bass array.


The Distributed Bass Array

The system I built is based on the Audiokinesis Swarm subwoofer system http://www.theabsolutesound.com/articles/audiokinesis-swarm-subwoofer-system/ with help from Audiokinesis himself, Duke Lejeune.

It uses four 10" Morel drivers https://www.parts-express.com/morel-ultimate-uw-1058-10-subwoofer--297-130

These drivers at the time included sealed 1.25 cf cabinets. These go together easily with minimal woodworking skill and tools, but that would mean four sealed cabinets. Duke suggested building two ported would give the added option of being able to easily tweak the bass response if needed by simply plugging a port or two.

With ported enclosures there is a relationship between volume, port dimensions, and tuning frequency. Basically, a bigger cabinet volume, longer port, and smaller port diameter yield a lower tuning frequency. A smaller diameter port doesn’t need to be as long. But it needs to be big enough to pass the volume of air quietly, without chuffing. Just a few of the many details designers like Duke know in depth. Me, I used one of the on-line port length calculators. Which just happened to give the same answer as Duke gave me. So I know the calculator was right.

As a result, two subs are 16" sealed cubes, and two are 16x16x24 ported cabs, with 3" diameter by 20" long ports. According to Parts Express this results in an f3 of 52 Hz for the sealed subs, while according to the on-line calculator the ported subs are tuned to 20 Hz. I take all this with a big grain of salt or as an estimated starting point at best, as room reinforcement at low frequencies is so big a factor. The goal is simply to get in the ball park, or close enough for a little port plugging, phase adjusting, maybe even a touch of EQ to do the rest.


The Amps

Duke’s Swarm system uses the Dayton subwoofer amp https://www.parts-express.com/dayton-audio-sa1000-subwoofer-amplifier-rack-mountable--300-811
While one amp would do nicely, because almost all bass is mono, still there is an advantage in terms of flexibility in using two. The Dayton amps are adjustable for crossover, level, and phase. A main reason a DBA works so well is the four source locations results in a smoothing of modes. Another factor is the way we perceive the size of the space we are in is related to the mixing or randomness of the lowest frequencies. The idea is that being able to adjust phase can further smooth or blend modes resulting in a greater sensation of being in a larger space.

At this point however I had a bit of a problem. My Melody 880I amp has no pre-out while the Dayton amps have only line-level inputs.

There are plenty of Line Out Converters but they all look cheesy to me. So instead I opted to convert one of my unused RCA inputs to a Line Out.

Converting speaker output to line level is a simple two resistor circuit. One 10k drops the speaker level, one 1k provides a little impedance to the outboard component. http://www.epanorama.net/circuits/speaker_to_line.html With a foot or two of spare silver wire and $2 worth of resistors soldered inside the Melody retains its stock look and is ready to drive the Dayton sub amps.


The Setup

This is the hardest part with the conventional approach, figuring out where exactly to put the sub to get smooth flat bass. Okay, technically not hard- impossible! With four however...

I spread the four around the room- near the left corner of the front wall, a little further than that from the front corner of the right wall, the remaining two on each side a little further back than the sweet spot. Fairly random. They all face the wall.

One run of ordinary 14 ga wire, at least for now. The two 8 ohm subs on each amp wired in series is 16 ohms, or 4 ohms in parallel. I tried both. The bass is a little more taut and defined at 16 compared to 4. Either way the result is leagues beyond a conventional sub.


The Sound

Finally! Setting up the Talon Roc the DBA replaced I knew would be a chore so for that I used a warbled test CD. Having read so much I was so sure the DBA would not be like that this time that I put on Bela Fleck’s Flight of the Cosmic Hippo. This track features some incredible low bass. Bass that trailed off to nothingness with my Talon Roc. Bass so low I had only heard it done even halfways decent one time- in the gigantic basement of a speaker designer (Linaeum) in Beaverton.

The DBA, speakers plopped down with zero effort, no cones, plain old lamp cord, and only the crudest guesstimate levels set, reproduced the Cosmic low bass of the Hippo so perfectly I could hardly believe it. Everyone talks about sub compatibility, matching, phasing, yada yada- forget about it. If my experience is anything to go by just plop four subs down, sit back, and enjoy. Try not to giggle. Seriously. Try.

The bass from these things is just crazy fast, articulate, powerful and smooth. Uncannily so. The Absolute Sound review above describes this so beautifully I have little to add. What I will add, as amazing as the obviously low powerful bass that you’re aware of hearing is, what is even more amazing is the effect of the very unobviously unpowerful low bass that you’re not.

What I mean is, several times now I have been listening to music that seemed to have no low bass content at all. Why then does it sound so much better? What is the DBA doing??? Sometimes, as with piano, it makes sense. The piano is a physically large instrument. The lowest A on a modern piano is about 27.5 Hz. With string bass its 31Hz to 41Hz, four or five string, depending. All likely with a physical presence that extends a bit lower than their fundamental notes. So with a little thought its easy enough to see why these sound better with the DBA. But what about the rest? Why am I getting the sense of the whole recording being bigger, better, more real and present in the room?

Or should I say, not that the recording is in my room but that I am in the recording room? Duke and others have talked about a sense of envelopment. Something I have experienced for a very long time going back to my first concert hall experience as a youth is the unique sonic signature of large spaces. If you’ve been to a concert you know the feeling. You walk in and not only your ears but your whole body feels the sensation of really, really, REALLY, I’m talking subterranean low bass.

Now honestly, the DBA is not that. Maybe with four 15" subs it would be. If its even on the recording to begin with. Not the point. The point is, the DBA is at least getting to the point where this is where you have to go to try and describe it. Its that good.
128x128millercarbon

Showing 9 responses by noble100

noble100's avatar
noble100
2,419 posts
 
millercarbon,

     Excellent post.  You not only detailed how to technically construct a custom  4-sub distributed bass array system and how you positioned/distributed the 4 subs in your room, you also did a very good job of describing the extremely high level of bass response performance that users of 4 sub distributed bass array systems should expect to achieve in their own rooms.
     As I know you're aware, I have been using an Audio Kinesis Debra  4 sub distributed bass array system in my room for about 4 years now.  The Debra is priced identically to the AK Swarm at $2,800 but with more rectangular subs than the squarer Swarm subs and both are good choices for those who don't want to build their own custom dba.  They're both complete kits that include 4 subs and a 1,000 watt class AB amp that also controls the crossover frequency, phase, volume and has a limited eq.  
    My main reason for posting is to state that the 4-sub dba concept has provided the best bass performance I've ever achieved in my combo 2-ch music/ht system and 23'x16'x8' room by a wide margin.  The bass is very natural, seems effortless and integrates seamlessly with my large (6'x2') Magnepan dipole panel main speakers. The system has the versatility and capacity to accurately reproduce fast, rhythmic, taut bass just as well as deep, powerful and impactful bass on both 2-ch music and ht content.
     Good bass response is generally considered to be the most difficult portion of the audio spectrum to reproduce in a high fidelity manner in typical homerooms.  As I understand it and you mentioned, this is mainly due to physics and the extremely long sound waves of bass frequencies. The lower the frequency, the longer the sound wave and these waves continue to bounce/reflect off room boundaries until they eventually run out of energy.  The direct and reflected bass sound waves inadvertently meet or collide at specific spots in the room at which points the bass is perceived as a peak, dip or null (bass is heard as exaggerated, attenuated or even absent). 
 Here's how the owner of Audio Kinesis, Duke Lejeune, has described the reasons and benefits of using a distributed bass array system: 
"The problem the Swarm addresses is, audibly lumpy in-room bass response. This is a fundamental acoustic characteristic of small rooms ("small" meaning "smaller than a recital hall" in this context). This room-induced peak-and-dip pattern will ruin the response of any sub no matter how smooth it starts out, and the Swarm is an acoustic solution to this acoustic problem.

The Swarm consists of four small subs that are intended to be distributed around the room. They are all driven by one central shelf-mounted amplifier.  The idea is, each individual sub inevitably generates a room-interaction of a peak-and-dip pattern, just like any other sub does. But these peak and dip patterns change dramatically with sub (or listener) location. So with the Swarm, you have four dissimilar peak-and-dip patterns summing at any given listening location.  

The net result is far smoother than any one of these patterns could have been without EQ, and this smoothness holds up much better at different listening positions around the room than it would with one or two equalized subs. And "smooth bass" = "fast bass", as it is the peaks in the response that make a subwoofer sound slow (the ear has poor time-domain resolution at low frequencies, so group delay is not nearly as bad as it looks "on paper")."

     I think this is a very good summary of the bass forces at work in a typical room and the reasons a 4-sub dba system is such an effective solution.  Based on my experiences with the AK Debra dba system, I can state without any reservations that the dba concept works like a charm in my system and room.

Enjoy,
 Tim
noble100's avatar
noble100
2,419 posts
 
     I think Duke has a knack of mentioning things that really illustrates how well a 4-sub distributed bass array system actually performs in a given room compared to 1 or 2 sub systems.  Things like, and I'm paraphrasing this, "Fast bass= Smooth bass.  Two subs are twice as smooth as one, four subs are twice as smooth as two and eight subs are likely good grounds for divorce". 
     Accurate, descriptive and kind of funny=memorable and effective, right?
     I find the most fascinating aspect to dbas for me is the psychacoustic dynamics involved in the how and why these bass systems work so well.  Things such as the fact that a single sub causes a large number of bass peaks and dips in any given room and the solution, counter-intuitively,  is adding 3 more subs at different room positions in order to cause a lot more bass peaks and dips to the room with the knowledge that the time domain is much less important on bass sound waves but the human summing of all the bass peaks and dips and the bass pressure levels in the room are.
     My experiences with the AK Debra dba in my system and room has also convinced me that running the subs all in mono mode, as opposed to stereo mode, has no detrimental effect on system imaging.  My perception is that the sound stage illusion has become, if anything, even more expansive, solid, stable and detailed. 
     I was a bit concerned, with the bass coming from 4 subs spread around my room, about whether the bass would integrate seamlessly within this sound stage illusion, that is that the bass was perceived as coming from the proper bass instruments positions within this sound stage illusion.  I was very pleased to realize that the bass produced by  the Debra dba was seamlessly integrated and I did perceive the bass as  coming from the proper bass instruments positions within this sound stage illusion.
     Since the source of bass sound waves below about 80-100 Hz is generally considered undetectable and my 4 subs were only set to operate below about 50 Hz, however,  I was a bit perplexed how I was able to perceive this low frequency bass as coming from specific spots (the kick drum in the center, an upright bass to the left, etc.) in this sound stage illusion in my room. 
     I've theorized that I'm able to localize the locations of instruments producing deep bass notes only due to the higher frequency harmonics or overtones of these instruments being reproduced by my main speakers and giving clues to the specific spot the related deeper fundamental tones of these instruments are originating from.  I'm assuming my brain is doing the important work of associating the fundamental bass tones with the associated harmonics and deciphering the specific locations of these bass instruments within my sound stage illusion.  All quite a bit of processing but the illusion sure seems very palpable and real.  
     I'm thoroughly enjoying these sound stage illusions on many, but not all, recordings and the added realism provided by the excellent bass response of the dba.  But I'm still uncertain if my theory is correct.  Any thoughts?

Tim
noble100's avatar
noble100
2,419 posts
 
Hello jbrrp,

     Below is a link to a thread I started awhile ago that describes the procedure I followed to first find the ideal position for each of the 4 subs in my room and, second, exactly how to wire everything up to the supplied amp.  
https://forum.audiogon.com/discussions/anyone-else-using-an-audio-kinesis-swarm-or-debra-distributed-bass-arra

     My room is a 23 x 16 x 8 ft living room that I also use for both 2-ch music and ht.  I have a 65" hdtv centered on the 16' front wall with a 6' wide and 2' high equipment rack centered below the wall-mounted tv.  I have Magnepan 2.7QR main speakers straddling the tv and a CC3 center channel speaker above it. My listening chair is centered on the 16' rear wall.
     For the subs, 2 are located at opposite ends of my front 16' wall with each about 2' away from the corner.  The other 2 subs are located more toward the rear of the room, with one along each 23' side wall, about 2' away from each rear corner and directly across the room from each other.  My listening chair sits between these 2 subs with each about 7-8' away from my chair. 
     So, I'd describe my setup as an "all around the room" sub placement like the placement Duke suggested for you.  I just followed the sequential sub locating procedure I described in detail on my thread linked above.  I had no preconceived idea where each sub would ultimately be positioned, I was determined to just follow the sequential procedure and trust that, if I did, the subs would be properly positioned in my room so that the 4 subs working in unison would provide the optimum dba system bass response performance in my room. 
      I had already decided I was going to position each sub in the room where it sounded the best and would redo the position of the furniture in the room if necessary rather than compromise the system's bass response performance.  Fortunately, I only had to move a couch a few feet forward along a 23' side wall to position one sub.
   
 I hope this helped,  
       Tim
noble100's avatar
noble100
2,419 posts
 
jbrrp1,

No problem, it really is very approachable. Funny, I even get excited about dbas if someone else decides to buy or build one.
It’s your choice, of course, but my opinion is that 2 Dayton amps can be beneficial since they’ll provide more reserve power for even better bass dynamics, more system flexibility for setting crossover frequencies, sub phasing and sub volumes and the potential to run the 4 subs in a stereo configuration.
However, I think you’ll be disappointed in configuring them for ’stereo bass’ for a couple reasons:
1. I don’t believe there’s such a thing as directional bass perception or ’stereo bass’ below about 100 Hz. I don’t know the exact bass Hz frequency that I perceive bass as directional, but I know it’s closer to 200 than 100 Hz.
To be fair and objective, though, you may be more sensitive to the direction of deep bass than I am. I’ve also never configured my dba system in stereo mode since I only have a single Dayton amp. But I do think it’s at least possible that higher frequency harmonics and overtones of deep bass notes, reproduced by the main l+r speakers, could provide clues to the location of the sources of deep bass fundamental frequencies.
Overall, therefore, I don’t think it would be wise to dismiss the possibility of perceiving ’stereo bass’ solely because it’s generally understood that bass is not directional below about 100 Hz. I believe this premise deserves further study, or perhaps I just need to further study the research and knowledge that may already exist on this subject. You don’t know what you don’t know, right?

2. Very few recordings, actually I’m aware of absolutely none, contain ’stereo bass’. I believe the bass is summed by recording engineers due to physical limitations on vinyl recordings and more out of convenience or traditional/common practice on cd recordings. However, with the significantly increased capabilities of current and potential future direct to digital recording techniques and technology, this may to lead to more recordings containing ’stereo bass’.

In my over 4 year experience of using the AK Debra distributed bass array system running in mono configuration, I’ve been more than thrilled with the bass response and consider it state of the art bass performance for 2-ch music and ht. It may be due to psycho-acoustic principles, but I perceive the mono bass as very well integrated, emanating from the proper locations and instruments on the sound stage and overall very ’stereo bass....ish’.
If progress does result in more content containing ’stereo bass’ in the future, I or you could always buy a 2nd sub amp and reconfigure the dba to run in stereo.

Your choice,
Tim
noble100's avatar
noble100
2,419 posts
 
Hello millercarbon,

     I like your experimental nature, always exploring and pushing the envelope.  You've become the mad scientist of in-room bass exploration.  I'm fairly sure you now wear a white lab coat when you're experimenting.
     I'm in the process of thinking of a good nickname for you, I'll let you know when the light bulb goes on.
     Very interesting how the Roc as a 5th sub helped so much. It makes sense according to Geddes (the more subs the smoother and better the bass) but the gain of going from 4 to 5 subs is expected to be marginal and not significant like you experienced.
     With the DBA concept working so well being based on using more subs to create more peaks and dips in the room, and our brains being relied upon to process all the resulting bass variances and average them out via psycho-acoustic principles to provide a perception of smoother overall bass response, it's a lot like making order out of chaos.  It makes sense to me in a macro sense but I have almost zero understanding of how this works in detail.
     Maybe the scientific explanation is you just got lucky.  My opinion is that we're very fortunate our brains are capable of making a 'bass smoothie out of a bunch of lumpy bass room modes'.  Don't bother trying to patent that phrase, buddy, I already did.  
     We're also fortunate Geddes, Toole, Welti, etc. took advantage of this brain peculiarity in processing deep bass tones and developed it into the black arts creation of the DBA concept.  It really does remind me of alchemy or an awesome magic trick.
     So, I'd suggest not trying to explain it and just enjoy it.  I'd be concerned I might sprain something important trying to make sense of this one.  I'm just going to refer to it as the Talon Roc Fifth Sub Mystery for now.


Later,
 Tim
noble100's avatar
noble100
2,419 posts
 
millercarbon,

     I think perception is reality in a practical sense.  If you consistently perceive the bass as better, then the bass is better.  It'd be nice if a few others perceived it, too, just to validate your sanity but it's not necessary.
     I'm thinking the Roc alone didn't work well previously in your system because there were bass standing waves causing bass modes at your listening seat. Now, with 5 subs, those aren't perceived and the Roc is freed to add its extra and more extended bass to the mix.
     I ordered some new white lab coats for you with 'Duke DBA Disciple' printed on the back.


Later,
 Tim
noble100's avatar
noble100
2,419 posts
 

     I was able to change the lab coat order but they were out of the Tommy Bahama silk lab coats.  So I got you 4 sets of their silk pajamas instead.  The pants have footies but I figured that'd be okay just around the lab, right?

     pineapple juice is already chillin'.


Tim
noble100's avatar
noble100
2,419 posts
 
Hello haywood310,

     Your 14 x16 ft room is on the small size and it's easy to swamp it with too much bass, especially with two HSU ULS15s. 
     I think the best solution for you is to sell the HSUs and get 4 morrell subs with the Dayton amp for about $2k.
     The 4-sub Swarm type DBA would be a better fit for your room and will be easier to configure and adjust the volume, crossover frequency and phase settings for optimum performance.  
     The 4-sub Swarm setup will actually make a small room like yours seem bigger, AK's Duke Lejeune says the smaller the room the better the DBA concept actually works.  The bass will be detailed, smooth and natural with plenty of reserve power for good dynamics.  
     The quantity of bass in your system/room will be completely adjustable via the Dayton amp's volume and crossover controls.  The trick will be maintaining some restraint to achieve, millercarbon's favorite, seamless integration with your main speakers.

Tim
noble100's avatar
noble100
2,419 posts
 
millercarbon,

     If he's going for maximum bass then he may want to use one or two of the HSUs.  I think four of the Swarm or Morrell subs with the Dayton amp would supply plenty of bass and headroom in his room, it does in my 23 x 16 ft room for music and ht.
     From my perspective, having some amplified subs like the HSUs does add more bass power, extension and flexibility as far as control and fine tuning is concerned but I find the convenience of controlling all 4 subs for volume, xover and phase preferable.  It's probably best haywood just is aware of his options and choose what he prefers.

Later,
 Tim  
More to discover