Subwoofer Reviews

As an audiophile, I've come to associate the size, weight, and price of a subwoofer as quick'n'dirty indicators of its quality. The subwoofers that have worked best in my large listening room—the Velodyne ULD-18 and DD-18+, Muse Model 18, REL Studio III, JL Audio Fathom f113, and Revel Sub30—each weigh more than 130 lbs and cost more than $2500. With some of my reference recordings, all of them have achieved what Robert Harley described in the April 1991 issue of Stereophile as the goals of a quality subwoofer: "seamless integration, quickness, no bloat, and unbelievable bass extension." Yet are back-busting weight, unmanageable size, and nosebleed cost essential to achieving those goals?

SV Sound doesn't think so. Their sealed, self-powered SB13-Ultra subwoofer weighs less than 100 lbs, yet boasts a 3600W peak amplifier. SVS sells this model directly via their website and offers buyers a 45-day, in-home trial period, with money returned in full if the sub doesn't work out. SVS's website has chat features, and provides Merlin, a subwoofer-setup wizard. Type in the name of your main speakers, and Merlin recommends the "SVS subwoofer that provides the best match, including the exact settings needed to optimize the sub's sound."

Intrigued by SVS's approach, I jumped at an offer by Nicholas Brown, SVS's PR representative, to review the SB13-Ultra.

The SB13-Ultra . . .
. . . is an impressively compact, sealed-box subwoofer. A 17.4" cube, it's 3.6" shorter, 3.1" narrower, 10.4" shallower, 63 lbs lighter, and $400 less expensive than the ported version, the PB13-Ultra, which I reviewed in the August 2008 issue.

Measurements show the ported PB13-Ultra has the more extented deep bass response, but SVS's Mark Mason noted that the sealed SB13-Ultra can take better advantage of room reinforcement of the very low frequencies; the ported version must use a steep subsonic filter to avoid overdriving its woofer below the port tuning frequency.

The SB13-Ultra has a single, front-firing 13.5" drive-unit. A custom-tooled, die-cast aluminum basket that holds the light, rigid Rohacell-composite cone with dual linear-roll spiders and a stitched, parabolic surround with large excursion. The motor, optimized with finite element analysis (FEA), consists of a bifilar-wound, flat-wire, eight-layer aluminum voice coil 3" in diameter, and a polyimide-impregnated fiberglass former with a custom gap-extension plate to increase its linear stroke, for lower distortion. The magnetic field is created by dual Genox 8H/Y-35 ferrite magnets, and the pole vent is oversized, for greater cooling. All of these components are configured in an "overhung" design that extends the voice-coil past the gap on either side of the pole-piece, to optimize its efficiency in a midsize sealed alignment. When played without limiters, filters, or equalization, the driver's low inductance extends its frequency response to 300Hz. Its manual states that it uses a "highly advanced and sophisticated Digital Signal Processor (DSP) . . . to achieve the target frequency response," and "features a frequency-dependent limiter/compressor algorithm with adjustable attack/release and compression parameters."

The SB13-Ultra is powered by a built-in Sledge STA-1000D class-D amplifier with an output of 1000W RMS (3600W peak dynamic). Featuring MOSFET output devices and a switch-mode power supply, the Sledge is smaller, more powerful, and more efficient than the 750W Switched Hybrid (class-A/B, class-D) amp used in the PB13-Ultra. Autostart and Green standby modes switch the amp on quickly when a signal appears at the input terminals.

Mark Mason told me that, using the CEA 2010 standard 31Hz signal in a 2pi environment with a microphone placed at 2m, at sound-pressure levels (SPLs) with less than 10% total harmonic distortion (THD), the SB13-Ultra's maximum peak acoustic output was 111.4dB, as compared with the 118.9dB claimed for the PB13-Ultra.

Simplified Controls
While the SB13-Ultra doesn't come with a remote control, the user interface consists of a small, rear-panel LCD screen and a single control knob, which SVS calls the Integrated Function Controller (IFC). Turning the knob scrolls through eight setup and control functions, each in turn displayed on the LCD. Push the IFC once to select a function, then turn it to scroll through the submenus. Quickly push it twice (double-clicked) to return to the top-level menu. The submenus include: multiple high- and low-pass crossover corner-frequency settings between 31 and 125Hz, plus two different filter slopes (12 or 24dB/octave); phase adjustable from 0° to 180° in increments of 15°; high-pass delay continuously variable from 1 to 10 milliseconds, to align in time the outputs of the satellite speakers and sub; three room-compensation filters (40Hz for rooms of less than 1400 cubic feet, 31Hz for rooms of 1400–2400ft³, and 25Hz for rooms greater than 2400ft3 (6 or 12dB/octave); two parametric equalizer (PEQ) bands offering 13 different center frequencies between 31 and 125Hz; and nine different Q values, from 2.0 to 14.4, for reducing the largest and widest room-mode peaks.

The IFC unclutters the SB13-Ultra's rear panel, leaving only: the unbalanced (RCA) and balanced (XLR) inputs and outputs for the right and left channels; a switch for selecting line or high voltage level; a power switch; and an IEC jack for the detachable power cord.

Room, Setup, Measurement
I've used the same listening room for over 20 years. Measuring 25' long by 13' wide by 12' high, it encloses a volume of 3900 cubic feet. The left wall has a large bay window covered by Hunter Douglas fabric shades. Under the solid-oak floor is an unfinished basement. Two area rugs cover most of the floor, including the space between the listening chair and my Quad ESL-989 speakers. Although large, the room's sparse furnishings allow these electrostatic panels to produce peaks of 90dB SPL at my listening chair. Through an 8' by 4' doorway, the rear of the room opens into a 25' by 15' kitchen.

The very first subwoofer I reviewed using this room was Velodyne's ULD-18, for the October 1989 issue. Accompanying a pair of Quad ESL-63 electrostatics, the ULD-18 did best when placed in a corner, and I used the same positions for this review. My Quad ESL-989s stood 6' 8" apart at their inner edges, the left speaker 18" from the left wall, the right speaker 18" from the built-in wall unit on the right, and both of them 5' 5" from the front wall. The SB13-Ultra was in a front corner, 3' behind the right-channel Quad. My listening chair was 7' 8" from the Quads' front baffles, and 10' 8" from the front of the SB13-Ultra.

Setting up, calibrating, and integrating an SB13-Ultra into an audio system is well described in the clearly written, 34-page manual, which recommends that the sub's room response be optimized either a RadioShack Sound Level Meter and Microsoft Excel, or the Avia II: Guide to Home Theater test DVD (Ovation B19485, $44).

Because I didn't have a A/V receiver though which to play Avia II, I used my Studio Six iTestMic, a professional-grade test and measurement microphone for the iPhone 4 and iPad. The mike plugs directly into the iPhone's 30-pin connector, and auto-calibrates while drawing very little power from the phone. It's far more precise than the iPhone's own mike for accurately testing and setting up subwoofers, as well as for measuring noise levels, and sound levels up to 120dB. Studio Six's AudioTools app runs the iTestMic, stores the data on the iPhone, and analyzes and graphs its measurements. For test tones, I played, on my Bryston BCD-1 CD player, a digital file of uncorrelated pink noise supplied by Kevin Voecks, of Revel speakers.

First, I ran the preamplifier output cables directly to a pair of Theta Digital Prometheus monoblocks, to run the Quads full range. Using AudioTools' Real Time Analyzer (RTA), their in-room frequency response measured 25Hz–20kHz (fig.1). This graph showed room-mode peaks at 80 and 40Hz, but the response fell off below 40Hz by 15dB at 25Hz.

I disconnected the Quads, attached the preamplifier's output cables to the SVS's inputs, and set the sub's output playing pink noise by turning the IFC knob until AudioTools' SPL module registered 75dB at my listening chair. I then ran balanced interconnects from the sub's high-pass outputs to the Theta amplifiers, and set the high- and low-pass filter corner frequencies to the recommended 63Hz with 24dB/octave slopes. I turned the Quads back on and adjusted the sub's output to match the Quads' acoustic output until the iTestMic RTA histogram bars were level at 100 and 40Hz. That completed the initial calibration and adjustment of the system.