Subwoofer speed is in the room, not the box

@clio09 Then I will say it. John Hunter is wrong. He is interested in selling as many subwoofers as he can and not in designing the best possible subwoofer.

@tomic601 ​​@erik_squires @phusis This is not all that complicated. Most of you are right, you just need to be able to explain it better.

The speed a woofer cone travels is a sinewave function. The average speed to produce a note at a given loudness is a function of size. The smaller woofer has to travel farther to displace the amount of air required to produce the note at that volume so it has to travel faster. The speed is not an issue but "farther" certainly is. Dynamic speaker suspensions are linear only for a very short distance. At some point, depending on the suspension's design, the suspension becomes progressively stiffer until it can not move any farther without ripping it apart. Because of this non linearity distortion increases logarithmically with excursion distance. This is why larger drivers or multiples of smaller drivers have lower distortion levels. They do not have to move as far so they do not have to travel as fast. Bigger is always better, not worse!

Boominess has two causes, the subwoofer running into the midrange and a resonant, shaky enclosure. To keep the sub out of the midrange people historically lowered the crossover point. This is a problem for a number of reasons. Much of the impact or dynamic factor comes at higher frequencies in the 80 to 100 Hz range. You want the sub running certainly up to 80 Hz and I will go no lower than 100 Hz. The solution to this problem is running steeper filters, not lowering the crossover point. The problem here is analog filters are terrible at this. You have to have a digital crossover then 8th order and higher is no problem. Running that high a high pass filter for the mains is mandatory or you will have a hot mess. This is advantageous anyway from a distortion and headroom perspective. 

Making an enclosure that does not shake or resonate is a very hard problem to solve. The easiest part is using a balanced force design. You put a driver in opposite ends of the enclosure running in phase. The Newtonian forces then cancel out. Making an enclosure that does not resonate is much tougher and very expensive, more expensive than most commercial manufacturers will tolerate because they have to remain competitive. This is the reason I make my own. You can see a picture of the final versions before finishing. I plan on doing a full pictorial of their construction  so others can copy them if they are inclined. These will be high gloss black. The walls are 1 7/16" thick. Plywood was used instead of MDF because it is stiffer. The individual sections are only 4" wide. The subs sit on the floor horizontally on two spikes right up against the wall. They lean on a special pad. Because they do not vibrate at all nothing will get transferred directly to the wall. Small sealed enclosures are always best with subwoofers but you have to have a lot of power and high resolution digital EQ. Then you can make any sub run flat as a carpenters dream. 

The last problem is time alignment and this is best done with a digital crossover and room control which is really speaker control. 

With subs running up to 100 Hz and cut off at 48 dB/oct or higher, if you want more impact just turn the sub volume up a little and you can have it without affecting the midrange or treble. I run mine hot by about 6 dB which gives the music a "live" feel at less than ear shattering levels. A great subwoofer system is more felt than heard. In reality, the faster a subwoofer cone has to move the worse will be it's performance. 

@audioquest4life  If subwoofers and main speakers are integrated correctly there is no reason to turn the subs up or down with any genre of music. A system that is tuned correctly does not care what genre you are playing. When I use the term , system I include the room in that category. 

Most audiophiles are ball parking it with their ears which are extremely poor calibration devices. There is no substitute for measurement.  

@audioquest4life I listen to everything from Nine Inch Nails and The Red Hot Chili Peppers to Cherubini String Quartets. I have eight 12 inch drivers in a 16 X 30 foot room, each one powered by 2500 watts. I am into chest thumping as much as anyone. My subwoofers are EQed up about 6 dB so I can get the live concert experience at more reasonable levels. They run up to 100 Hz and are cut off at 48 dB/oct. I never change settings for any type of music and I can thump your chest into the next state and soar with The Lark Ascending, all on the same settings. The music has the choice of how it wants to sound.

@erik_squires Excellent! Speaker specs are worthless. Amplitude curves taken at 1 meter mean almost nothing. A speaker that goes down to 30 Hz at one meter might make it to 60 Hz in real situations. A speaker that is more omni direction is going to sound brighter, even sibilant in a room as compared to a directional speaker with the same specs. It is more important to understand what the design of the speaker is capable of and how to choose one that fits your needs and situation. Most people choose a speaker that looks good or suits their wife's sensibilities paying little attention to speaker design. They might look at the specs and if they do misinterpret them. Not all speakers are designed intelligently and some are awful.  

@phusis I have no experience with horn subwoofer. For most of us they are impractical do to size constraints. For sure distortion will be lower for any given size driver due to efficiency. They problem is low bass will rattle and resonate almost anything. IMHO is is much easier to make a small enclosure resonance free with clever design and balanced force construction. It will not come remotely close to the efficiency of a horn but is way more practical from a size perspective. You are doing exactly as I suggested for crossover and slope. The game is keeping the sub out of the midrange or you will have mud. You are running 85 Hz @ 36 dB/oct. If you move up to 100 Hz you will have to steepen the slope. I can change crossover points and slopes on the fly which is very helpful for AB comparisons. 

@audioquest4life  The largest I would go is 15". The reason is the larger drivers have more trouble maintaining pistonic motion, they wobble. If you want more subwoofer use multiples of smaller drivers. Two 12's make a 15. Two 15s make an 18. So, go with four 15s. I use 12" drivers because the 15's would require a larger enclosure which will not fit in my situation and larger enclosures are more difficult to control from a resonance perspective.

@gdaddy1 Rel is only interested in selling as many subwoofers as they can. Their method of implementation is worse than silly. 

@benanders The solution to your problem is digital bass management with room control, which is really speaker control. You can change the volume of the subs instantly with sliders for each individual channel on the computer. Room control will even out the frequency response of the sub with the best response at the listening position. I advance the volume of the subwoofers by 6 dB. with the crossover point and slope I use this gives live recordings the thump of the real experience at less than ear damaging levels and I do not change it at all for any given recording. These are choices that the mastering engineer makes and who am I to alter his art? 

Digital bas management uses measurements you make of your system. It sets the crossover point and slope, and aligns the subs with the mains in time and phase perfectly. After the system is set up it will be flat and your only adjustment by ear or feeling is relative volume. Low bass you feel more than hear. Go to a live jazz club and feel the bass and drums. That is what you want at home without being able to identify the subwoofers by ear. If you can hear the subs as individual drivers you need to rethink your integration. 

@gdaddy1 Subwoofer manufacturers are in business to sell subwoofers PERIOD.

They could care less about the performance of your system. Rel is handily the worst but the vast majority of them do the same half baked thing and supply their subwoofers only with low pass filters. Their reasoning is if they make things more complicated and expensive nobody would by their units. The sad thing is they are right. Digital subwoofer management has been around since 1995, invented by Radomir Bozevic (the Boz of TacT Audio). Some subs include a watered down version of room control but without a high pass filter they have no control over the main speakers which is 50% of the issue. Proper bass or subwoofer management has to know what the main speakers are doing to mate the two correctly in time and phase. Then there is the marked improvement in main speaker performance when you relieve them of the lowest octaves. This can only be done by direct measurement of a system in it's own environment. Any other way is wishful thinking which you seem to be very good at. 

There is, at this time, only one best way to integrate subwoofers into a system and that is with Digital bass management. There are now units available from MiniDSP and Anthem that are very reasonably priced. Benchmark Media Systems uses a MiniDSP SHD Studio with two of their DACs to run their own system and they get great results with it. (Dirac Live) For people who can spend more there is the Trinnov Amethyst and the DEQX units which will be available to the general public shortly. The Anthem STR preamp is 1/2 the price and a great performer.

@clio09 One of the best thing about digital signal processing is you can add subwoofer and eliminate the "bump". Full range speakers like ESLs benefit the most from a high pass filter. 

@clio09 I did not know anything about Roger until I bought some of his tubes, the best I ever had. I use Sound Labs speakers and before that Acoustats.  I have been crossing between 100 and 125 Hz for 30 years. It requires a very steep slope to keep the sub out of the midrange. The reason so high is that full range speakers are very susceptible to Doppler Distortion. I chose 100 Hz because that is the point that I could not see the diaphragm move. Aside from lower distortion I get another 10 dB of headroom. ESLs will go extremely loud if they do not have to make low bass and they don't have to force an amp to drive a 30 ohm load.

@clio09 Again, Roger is correct. Power and damping drive subs the best. But the real difficult aspect of ESLs is the very low impedance way up high. My tube amps have an output impedance of 1.75 ohms, but my speakers drop down to 1 ohm at 20 kHz. If you look at the system's amplitude curve, it starts rolling off at 12 kHz  and by 20 kHz it is down 60 dB! I just got my DEQX Pre 8 digital preamp and the plan is to bi amp the transformers. I am going to use a Bricasti Design M25 to drive the high frequency transformer. The Bricasti has a very low output impedance and will easily take the speakers up to 20 kHz. It also has a huge power supply and will continue doubling down to 2 ohms. Sonically it reminds me of my old Krell KMA 100s. I ran those till they self destructed. Roger West recommended a crossover frequency of 5 kHz, so I will start there. In an interesting inversion the tube amp does a stunning job of driving the lower frequencies and would happily take the speakers down to 24 Hz, If I let it. Solid state amps suffer because of the high impedance, the MA 2s do not. 

@gdaddy1 That is true, the first motive of any company is to stay alive and to do that they have to sell product. The problem with REL is it seems to be the only motive. Anyone can shove drivers into a box and make up a marketing story. If you want really bad bass and a lot of distortion follow Rel's instructions. I have been using subwoofers since 1978 with electrostatic speakers and have been through every possible permutation. Commercial subwoofers are so bad I design and build my own. Randy Hooker made the best subs you could buy, brilliant design, but way too big for modern sensibilities. Mine are the best subwoofers you can't buy, but unlike Hooker's subs you need a lot of power to drive them. I took me 30 years and 3 trials to get them right. You can see pictures of them in unfinished form at the link below. They are finished in high gloss black polyester and clear coated. If you have questions about the design fire away. There will be a full pictorial of their construction in case anyone wants to have a go at it. 

@phusis Thanx!

I think it is more a matter of, you have this idea in your head and dam the torpedoes you are going to do it regardless. I certainly am that way. Each enclosure has 84 individual pieces. I have a total of 4 coats of polyester lacquer to spray, wet sanding between coats and the last coat has to be sanded 4 times to 2000 grit then polished two times. I am putting together a pictorial diary of their construction in case someone wants to give it a go. I am not making any more for any reason ever. 

I have a much different situation than you. My main speakers are line sources all the way down to 1 Hz. The subwoofers, in order to match the volume at increasing distance have to act like line sources. I achieve that by spacing the drivers at the right interval right into both side walls so that they are acting acoustically as one driver, a bass line source array. No matter how large and powerful you make a horn it is still point source unless you space one every four feet from wall to wall. Instead of one very efficient driver in each channel I use four 12" drivers in each channel and 2500 watts per woofer. The over all distortion at any given volume below 100 dB (already too loud) is probably about the same. You do not see horns at big concerts any more. They hang two 40 foot curved line arrays and  A LOT of WATTS. Who cares about electricity bills? 

@100Hz and 36 dB/oct you will only be 20 dB down at 200 Hz. Imagine a phono stage with a signal to noise ratio of 20 dB. using 48 dB per octave you will be 60 dB down at 200 Hz which is low enough to be masked by louder signals. Other drivers are better at doing midrange than big woofers, that's just life. With ESLs there is a stark difference. If the subs run into the midrange you will easily be able to localize them which to me is very annoying. The subs are integrated correctly when the low bass is there, more felt than heard and you would swear there wasn't a sub in the room. 

@phusis The Sound quality at stadium concerts is in general awful, but that is not the point. The line arrays project power better which is what it is all about at these concerts.

Exactly, the horizontal array of subwoofers forms a horizontal line source. Who says a line source has to be vertical?

If the subs are stereo, they are matched to the main speakers in time and phase, and the volume is set correctly you can not localize a sub running up to 100 kHz rolled off at 48 dB/oct. If you could I would not be using that combination. 

There is no such thing as too much head room:-)

There is nothing wrong with a proper point source system. The image will be smaller and the acoustic power will roll off faster with distance. All other aspects of sound quality are the same. Many people prefer the point source presentation. They do not like sitting up front.

That is the usual complaint with ESLs, they are too polite and if you make them do low bass you are correct. THEY HATE MAKING BASS! Which is why I got into subwoofers back in 1978. I had Acoustat X's and they were way too polite. Taking 100 Hz and below away from them unleashes a lion. With an 8 foot tall line source ESL in a room with an 8 foot ceilings they are more powerful than most horn systems, but they do not go as loud. I can get up to 105 dB and that is it. The ESLs do not give up. I run out of power. By more powerful I mean things like snare drum snaps and bass drum kicks have more visceral content. The problem for ESLs is, to get reasonable efficiency the stators have to be as close to the diaphragm as possible and the diaphragm has to be stretched tight. There is no suspension. Thus, an ESL does not have enough space or compliance to make low bass at volume. The diaphragm excursions are too long for them. But, take 100 Hz and below away from them and you can not see the diaphragm move at all. The only limiting factor is transformer saturation. The Sound Labs have two transformers separated by a 6 dB/oct xover at 5 kHz. I am going to remove that crossover and use another amp and a digital xover to run the high frequency transformer. 

The major benefit of both our systems is the controlled directivity of both horns and dipole line sources, minimizing room interaction which produces a much finer image. Both have ultra low distortion. Because of their efficiency, horns will go louder and require much less power to do it. The ESLs will produce a larger image moving you to the front of the venue. All roads can lead to Rome if close attention is given to every aspect of performance from cartridge alignment to matching the amps correctly to proper control of room acoustics. It is much, much easier to obtain great results with a digital preamp in both systems. Having complete control over frequency response allows one to match the channels perfectly obtaining the best image obtainable. Digital crossovers are superior to analog ones and jacking the bass at 6 dB/Oct below 100 Hz produces lifelike results at reasonable ear saving volumes.