Subwoofer: should we even use them at all?

here's the thing about subs. you have to put them in a position to succeed.

what do i mean by that? first off, older subs are too slow. integration issues were rightly noted. thats not the case with the new 'super-subs' that are out there (JL Audio, REL, Velodyne, etc). these newer subs are very quick and can keep up and properly augment mains in the right circumstances, leaving a drastically improved playback for those who will put them in a position to succeed.

what circumstances are those?
1) sealed subs (no ports)
2) sealed mains (no ports)
3) sub position flexibility (but close to the front plane of the mains)
4) adequate setting flexibility (namely, LP filter freq and slope, phase and volume)
and finally
5) digital room correction within the subs (JL Audio has the ARO system).

and here's why:
one of JL Audio's designers wrote a white paper in which he astutely identified that each source of bass will make the integration of a sub more challenging. ports are a source of bass, and a non-linear one at that! main channels are typically designed using ports to make the speakers A) more dynamic B) louder and C) seemingly deeper in the bass (though its really just a port-tuning phenomenon which tricks the listener. the truth is that the frequencies roll off sharply below the port tuning frequency). ---the nonlinearities of the port (whether on the mains or the sub) create a host of problems (room nodes, freq bumps) that are tough to work around. ergo, #1 and #2 above...the fewer ports, the fewer sources of bass that you have to contend with. --this is reason #1 sub integrations fail.

positioning is a function of getting the timing relationships correct. subs tucked way behind the plane of the mains means their sonic impulse has to travel further to get to the listener--and this gives rise to timing errors. plan on putting the subs close to the mains to maintain timing relationships. this is reason #2 sub integrations fail.

setting flexibility is critical to getting the most out of your subs. when you start with sealed mains, you have speakers that will have a very linear rolloff in the bass (excl room nodes, but more on this later). the subs come from the bottom up...where you have to play with settings to get the subs to roll off quickly once you reach the frequencies that the mains handle properly. if your speakers roll off 3 or 6db at 50hz, then its natural to start the subs LP filter around 40-60 (experiment) and play with the slope (12, 24 db per octave) to get the subs to drop off quickly so as not to intrude on the mains. phase should be set before anything though---check phase w/ mains and dial in for least output when both mains and subs are running, and then reverse phase of the subs. now that phase, crossover and slope are all set, play with volume to your liking. ---improperly setting phase, x-over, slope & volume (with volume the least important factor) is reason #3 sub integrations fail.

(note the mains are running full range---no parts in the signal path is best, and its preferred, given sealed mains and their natural linear rolloff).

finally, some digital room correction is in order. most high end subs have some feature to do so, and generally they're about removing room nodes (points of elevated volume)..these are far more problematic than nulls (points of reduced volume). run the sub's digital correction which will identify the nodes and shelve these frequencies down in order to reduce the node...and the beauty is, with sealed speakers / subs, you only have room-generated nulls, not port-generated! the digital correction will eliminate the node for the sub, so even if the main still aggravates the room-node, it will be counteracted by the sub's digital correction which has nulled out this freq from the sub's playback.

i've done just this, using magico S3s w/ JL F110s, and the setup is startling---the above is best bang for the buck sonics possible (as big full range speakers are expensive, for one reason---bass is expensive! so save a bunch of $$ and get the bass into a diff box and put the subs in a position to succeed).

then again, YMMV.

cool post...thanks Rhyno.

Richard Vandersteen advocates the use of powered subs. Not only does it extend the bottom, but cleans the mids and highs as well

Rhyno,

Informative post indeed.

Rhyno,

While there's more than a little good info in your recap of JL's white paper, there's also some misinformation - not sure if it's yours or theirs.

Your description of port behavior may get the spirit of the design's performance more or less right, but it's not exactly correct: Ported enclosures are not merely a "trick" resulting from a port tuning phenomenon. A properly tuned ported box will drive the 3db down frequency lower from any given driver/suspension and will roll off faster below that point than the same system rolls off in a sealed box. That's more extended bass - no trick.

Bear in mind that, as you move lower from the tuning frequency, there will eventually be less bass output from the ported box, due to the more rapid roll-off below the tuning frequency. However, if the port tuning is low enough, as it is on many high-end ported subs (I believe most or all of the SVS subs can be tuned below 25hz, for example), that may not be an issue. For music, the rapid roll-off of a ported sub tuned this way will almost never be an issue.

However, there are definitely other trade-offs with a ported box and I personally do agree that it's much easier to get a good sounding set-up with sealed subwoofers than with ported designs.

As to positioning subs in/near the same plane as the mains, many of the external bass management/digital room correction (DRC) systems will delay the electrical signal to either sub or mains to account for the delta in distance between the sub and mains. This allows more flexibility in optimizing position of both the subs and mains, which may be optimized when they are far from the same plane.

I'm not sure whether JL's own system offers this, but I'm guessing that it doesn't. Since JL's products offer in-sub room correction, it shouldn't surprise that they suggest that it's where DRC belongs (and it also explains their recommendation on room positioning). Personally, I use an external bass management DRC system and enjoy the added placement flexibility.

Running the mains full-range may offer the simplest signal path, but it also eliminates one of IMO the biggest benefits a subwoofer offers. All drivers produce more distortion as frequency drops - the longer driver excursions required for lower frequencies will reduce linearity (increase distortion). The good news is that a high quality sub (like your JL which is among the highest quality IMO) is better equipped to handle the heavy lifting at low frequency than is the woofer in virtually all main speaker systems (even including your Magico, I'd think). If you actively low cut the mains, you remove the heavy lifting from their woofer and shift the burden to the subwoofer, where it belongs. Your mains will benefit from the narrower bandwidth they're being asked to handle. So, there's a trade-off; simplicity vs optimizing bandwidth to driver. Some may prefer the full-range option (I definitely don't), but that's a matter of personal preference not system optimization.

Also, you'd need to consider the room-correction side of the issue. If your DRC is in-sub, I assume that it's functional only to the sub's high cut frequency, yes? If so, and you cross the sub out at 60ish hz, you're only room correcting to that point. In every room that I've ever measured, serious response irregularities run up to 120ish hz and significant irregularities persist above 200hz. Below about 80hz, passive treatments become increasingly cumbersome. If you limit your DRC to 60hz and below, there's a lot of room clean-up that you're foregoing.

At the end of the day, I'd say that most of JL's advice (at least as you've characterized it) is sound, but I'd also note that it's definitely slanted in favor of selling their products.