Sealed subwoofer for ESL-63

There actually are dipole subwoofers---the OB/Dipole Gradient offered in the 1980’s for the QUAD 63, and the OB/Dipole now offered by Rythmik/GR Research. It consists of a pair of 12" OB-specific woofers mounted on an H- or W-frame---an open baffle, dipole sub with the Rythmik Servo-Feedback system, the only such sub in the world.

Ya’ll know Duke is a subwoofer expert, and his statement that dipole subs have smoother in-room bass than monopoles is, contrary to shadorne’s claim that "at these low frequencies they behave as an omnidirectional source"---an incorrect and mistaken notion, absolutely and incontrovertibly true.

A dipole sub has a null to either side (just like dipole speakers), propagating bass and bass-loading the room in one less dimension than an omni---not the width of the room. Consequently, fewer room modes are excited---less bass boom. That is an advantage inherent in dipole subs, and dipole speakers as well.

An OB-type dipole sub has one additional benefit---no sealed or ported enclosure to itself be a source of resonance or boom. As I described above, an OB sub has it's woofers are mounted on an open baffle, not in an enclosure. Lean, clean, and pristine!

Let me offer some clarification to my earlier post.

My statement that a dipole has smoother in-room bass is based on a paper written by James M. Kates and published in the Journal of the Audio Engineering Society in 2002.

It takes the ear a long time to hear low frequencies, relatively speaking. We cannot detect the presence of bass energy from less than one wavelength, and we must hear several cycles (several wavelengths) before we can detect pitch. If you stop and think about how long these wavelengths are, you’ll see that by the time we can hear low frequencies, they have already bounced around our little rooms quite a bit. So the room’s effects are all over the sub’s output by the time we hear it. In other words, perceptually we cannot separate the subwoofer(s) from the room - they form a system, for all practical purposes.

The idea with a distributed multisub system is, each sub inevitably generates a nasty room-interaction peak-and-dip pattern. But because the subs are intelligently scattered, these peak-and-dip patterns are all significantly different. The sum of these multiple dissimilar peak-and-dip patterns is pretty darn smooth, assuming we started out with four subs. This smoothness holds up pretty much throughout the room because at any location within the room, we have the summing of four dissimilar peak-and-dip patterns. And since smooth bass is fast bass (both literally and perceptually), the net result blends very well with dipoles.

Duke

Jzzmusician wrote: "I’ve got a pair of Quad 2805’s. I love pretty much everything about them but I would like a bit more bottom end... What would you recommend for a subwoofer(s) in the 2-3k range?"

I’ve owned three pairs of Quads and many pairs of SoundLabs, so I’m somewhat familiar with the superb pitch definition of a good dipole bass system. You can hear every little nuance of what the bass player is doing.

In my opinion, the most promising approaches to adding deeper extension to your Quads are probably four small monopole subs, and two dipole subs. Both approaches have similar smoothness across most of the bass region. Briefly, the more bass sources the better, and each dipole is like two bass sources, from an in-room smoothness standpoint. In both approaches, the bass energy from all these different sources bouncing around the room combines in semi-random phase to produce much smoother response than you’d get from a single sub.

Obviously two dipole subwoofers are gong to blend well with two dipole main speakers. But four small monopole subs, scattered around, have about the same in-room smoothness as two dipoles, so they also blend well with two dipole main speakers.

But there is an important difference:

At the bottom end of the bass region, where the wavelengths typically become quite long in relation to the room’s dimensions, a multiple monopole system would tend to have a rising response, because at these long wavelengths their outputs are effectively combining in-phase. And in-phase combining results in 3 dB more SPL than semi-random-phase combining.

On the other hand, at these long wavelengths, the in-phase and out-of-phase energies of the dipoles will combine to produce a falling response, because they are summing towards complete cancellation.

Neither is ideal. But the little multi-sub system has a trick up its sleeve: You can reverse the polarity of one of the subs! This actually results in smoother response over most of the bass region, and then no fat-sounding hump down at the bottom end of the bass region. Smooth bass is "fast" bass, and the net result is smooth bass that extends lower than a comparable dipole subwoofer system would go. And subjectively, the distributed multisub system has the impact that dipole systems lack. So arguably best-of-both worlds: Articulation and pitch-definition competitive with a good dipole, but with impact too.

Yeah I got a dog in this fight... four small dogs, to be precise. So feel free to take my comments with as many grains of salt as needed.

Duke

dealer/manufacturer/distributed multi-sub guy