October 2001

Ever since I started to experiment with amplified subwoofers -- both via line- or high-level connections -- I've been surprised to discover how the addition of real sub-bass affects the entire audible frequency spectrum all the way up into the treble. The first thing I've learned is that unless speakers require filtering, I much prefer a high-level subwoofer connection. In this hookup, the subwoofer derives its signal from the system's main amplifier. Its sonic signature is thus passed onto the sub. Using the more popular line-level connection from a pre- or sub-out circumvents the main amp. Especially when it's a tubed device, this tends to cause a sonic discontinuity between tube mid/highs and solid-state bass.

Supplemental materials by British subwoofer manufacturer REL first hipped me to this phenomenon years ago. I was auditioning a Sonus Faber Concerto/REL Strata II combination during a weekend home trial. Following REL's advice, I compared both hookups via my Mesa Tigris tube integrated. I found the speaker-level connection to be more coherent and better blended -- the additional sub was rendered invisible as a separate sonic contributor. Since then, I've auditioned Audio Physic's Luna subwoofer, Hsu's VTF-2 and Soliloquy's S-10. Though my earlier push/pull loyalties have shifted to SETs, I still prefer this subwoofer hookup scheme.

For biamp scenarios, popular notions suggest that a muscular high-current amp should control the bass with an iron fist while a sweet and dainty tube amp may romance the higher frequencies. While this does hold a certain theoretical appeal, closer inspection reveals that it's more likely to cause grave timbral discontinuity. Biamping instead with an active subwoofer (connected to the main amp driving a two-way speaker) seems a more promising approach to retain tonal coherence.

Tonal balance as seesaw

A speaker designer once told me that bass-limited monitor speakers enjoy a more realistic tonal balance when their top-end is slightly rolled off. This immediately made sense when I envisioned tonal balance as a seesaw centered at approximately 1000Hz, between middle and upper midrange. This central fulcrum perfectly balances the aural playground plank if we imagine the ten octaves of human hearing divided into two groups, with the five lower octaves sitting on the left end and the remaining five on the right.

Let's tell the two lowest octaves to take a hike down low-rider alley, shall we? This lightens the scale on one end and would occur with speakers whose frequency response was 80Hz to 20,000Hz and thus didn't capture the 20-40Hz and 40-80Hz octaves. Our seesaw is now heavy in the treble, with the macho bass boyz suspended in the air begging to be let down. The speakers sound predictably lightweight, thin and bright. Their center frequency has moved to about 2000Hz, shifted up by one octave, which is half the difference of the two missing octaves. How to restore normal tonal balance? Logic suggests removing the two uppermost treble octaves. Rolling off our fictitious monitor at 5000Hz should offset the two missing bottom octaves.

In reality, that's not necessary. Most acoustic music features very little content in the lowest octave below 35Hz. If a bass-shy speaker with 80Hz bass extension introduced a moderate roll-off at 10,000Hz, its tonal balance would sound less top-heavy on music that didn't extend much below 40Hz. This speaker would simply lack the lower octave, be slightly recessed in the highest octave but sound reasonably balanced overall.

Another law of relativity

It's easy to appreciate how the mere absence of bass directly affects our perception of the remaining frequency spectrum. Conversely, adding to our two-way monitor an appropriate subwoofer with the proper low-pass slope and gain setting will significantly alter the monitor's perceived tonal balance without causing any measurable changes in its frequency response.

To illustrate this effect with a visual example, think of three identical photo prints that depict a red rose in a yellow vase. Frame each print with a differently colored mat -- white, black and blue. Even though you know the color values of the three roses to be identical, the blue mat will significantly affect how your eye perceives it. It will not look the same as the others. Tonal balance likewise relies on what it is framed by -- bass and treble extension. Without changing measurable midrange response, adding or subtracting bass and treble appreciably impacts how we perceive the whole.

Inaudible bass that's audible elsewhere

It turns out that this truism extends beyond tonal balance. While there may not be much regular music signal below 40Hz, the addition of a subwoofer makes a very obvious difference even with material like female vocals (perhaps accompanied by piano and violins) that lack true bass. A superior subwoofer enhances ambience, realism, treble sweetness and midrange presence. Ambience is said to register with the more reptilian and primitive part of our brain that's a remnant of survival skills early hunters relied on. At least partially, it seems triggered by subtle low-frequency pressure changes that our ear/brain mechanism translates into the size of the recording venue, its reverberant character and the relative location of the performers within it. (Scary moments in movies are nearly always accompanied by extremely low high-pressure sounds that are more felt than heard to stimulate immediate and involuntary bodily responses of dread or alertness. Our senses register presence.)

Bypassing a subwoofer in an A/B comparison reveals that despite the obvious absence of regular bass data on solo vocal tracks, something is present. When not reproduced, it causes a reduction of stage depth and air that's accompanied by the dissolution of dimensional richness. While not turning two-dimensional per se, the subwoofer-less presentation does lose a certain spatial expansiveness and distinction.

I don't purport to understand enough about this subject to do more than speculate about what these "ambient data" might consist of. A compelling argument I've heard mentions subharmonics. The harmonic progression of overtones above a tone's fundamental frequency also creates mirror images below the fundamental. These subtle sounds are called sub-harmonics (or undertones) and occur multiple octaves below the main frequency. If this theory about ambient cues were valid, it would be easy to appreciate how fundamentals below 100Hz generate sub-harmonics that reach well into true sub-bass territory and thus require either a genuine full-range speaker or a subwoofer to be rendered audible.

We know that the harmonic structures of instruments and performers act like aural fingerprints. The more accurately a system captures them, the more realistic instruments sound -- to the extent that an aficionado can distinguish a Bösendorfer from a Steinway and a Guaneri from a Stradivarius. It makes sense how the added dimension of reproduced subharmonics would present our hearing mechanism with vital data that distinguish artificial (or reproduced) sounds from the real thing and thus enhance realism when present.

Damnation and redemption

How powerfully changes in the bass range impact the overall sonic presentation was recently driven home again with the Avantgarde Duo 2.2 horn speakers. They currently have me hope -- in vain, of course -- that the manufacturer might forget whom he sent 'em to. As this month's "Earmarked!" details, very minor changes in the gain, subsonic filter and low-pass settings of their actively amplified dual 10" subwoofer modules instill potent changes of tonal balance. But it goes further than tonal balance. A recent faux pas serves to illustrate the point again.

I had, on a lark, decided to revisit my alternate digital front-end. It straps the digital out of my Marantz professional CD recorder to the outboard resolution-enhancement and DAC boxes of Perpetual Technologies' P-1A/P-3A duo. With the resident Cary CD 303 player temporarily displaced, insertion of the first CD caused my ears to knot up in dismay. Aurally, I experienced the where-are-the-scissors anxiety of a bad hair day: the airiness of the upper frequencies had vanished into, ahem, thin air. This rarefied "lack of oxygen" suffocated the fabulous three-dimensional transparency of the sonic fabric and crippled the palpability of the performers in the soundstage. Openness and intelligibility congealed; something darker, opaque and less distinct took its place. Vocals sounded thicker and constricted, bass predominant and overdone like mushy pasta.

Merde, thought I in best high school French. You sorry artiste, couldn't you leave well enough alone?

Then I recalled that one of the Perpetual duo's strengths was bass slam and weight. Substituting the digital front-end probably upset the carefully dialed-in balance of the whole system. Thankfully, temporarily owning actively amplified subwoofers as part of the main speaker system prevented the usual frantic reactions -- putting half my components on Audiogon in an insane poker move to replace them. Instead, I simply changed the subwoofer gain and low-pass settings. Voilà, the magic returned in an instant.

Are conventional full-range speakers really that desirable?

I was only half gone with the Audiogon comment. Anyone with integral full-range speakers lacking bass adjustments knows about the trade-offs between siting them for best disappearing act versus obtaining the most even, lump-free bass response. Consider, too, the disproportionate expense of extending a speaker's bass response from 40Hz to 28Hz (never mind 20Hz). Suddenly the monitor-subwoofer brigade appears not only like the smart money. On a whole and when considering real-world living rooms rather than dedicated, treated sound rooms, they probably also enjoy better sound -- for less, and in a more attractive package.

Basso profundo indeed. There's a lot more to bass extension than stomping dinosaurs, rocket launches, 32-foot organ pipes and raves. Consider adding a (good) subwoofer to your system!

...Srajan Ebaen
srajan@soundstage.com

• The Y-Files Current Issue