Revel Performa F30 surround speaker system:
Page 2 The S30 surround can be used as a dipole or monopole speaker. It can also be wired to switch between these options by means of an external 6.8-12V DC trigger. A three-position tweeter-level control (-1, 0, +1dB) is also provided, as is a single set of speaker terminals. When switched to dipole mode, the two 4-inch drivers, mounted on opposite angled side faces of the cabinet and wired out-of-phase, provide all of the S30's output above 250Hz. In monopole, these two drivers are silent and the S30's 1-inch dome tweeter reproduces the highs.

The S30's high-frequency extension is curtailed in dipole operation. By just how much, Revel does not say—frequency-response specifications are provided only for the monopole mode—but a 4-inch driver cannot have the treble response of a 1-inch tweeter. It could be persuasively argued that dipole operation emphasizes envelopment, and that extended high-frequency response is unnecessary. Nevertheless, when I mounted the S30s on the sidewalls of my home theater, just to the rear and above the listening position, I almost invariably preferred the monopole mode, which I used for most of the listening tests with both music and soundtracks. In the past, I've been a dipole person. It's too early to say if my new home-theater room will permanently change that preference, but it's nice to have both options.

Sweet and Low Down
In many ways, the B15 subwoofer is the most interesting speaker in the whole Performa line. I've already mentioned its prodigious ratings for bass extension and output. In addition, its lowpass and highpass filters offer unusual flexibility. They are adjustable for slope (12 or 24dB/octave lowpass, and 24 or 48dB/octave highpass) and crossover point (30, 40, 50, 60, 70, or 80Hz). Dissimilar low- and highpass settings can be used, and the phase of the lowpass filter is continuously variable from 0° to 180°. The lowpass filter can also be turned off, should you choose to use the crossovers in your surround processor.

But there's more. A few high-end subwoofers provide built-in equalization to assist in tuning the bass to the room, but the B15 is the first to my knowledge to incorporate three bands of parametric equalization. At any point from 18Hz to 80Hz, each band can put a notch in the frequency response as narrow as 0.1 octave and as deep as -14.5dB. Thus you can tune out the three worst room peaks in this frequency range, or use more than one of the equalizer bands for a particularly nasty boom. Each equalizer also provides positive settings of up to +6dB, but attempting to eliminate room bass dips by boosting the response is not generally recommended. More than likely, that dip is a room null—an acoustic “black hole." Boosting at that frequency will gain you little and may seriously overdrive the subwoofer.

Setup
The F30s were positioned directly to the left and right of a 7-foot projection screen in my 15.5x26x8-foot home-theater room. Because the room is relatively narrow, and had to also accommodate the screen, they were about 2 feet from the sidewalls—a bit closer than I would usually prefer and closer than they would be in a traditional 2-channel stereo system. I toed them in so that their axes crossed just in front of the seating area. The C30 center was mounted on its optional stand just below the screen, and the surrounds, as already noted, were placed on the sidewalls just to the rear of the listening position and slightly above ear height.

One aspect of the S30 surround speaker's design, no doubt intended to assist custom installers, will likely annoy do-it-yourselfers. The mounting arrangement uses flush-mounted brackets in the cabinets, which fit into shoulder screws driven into the wall. (A shoulder screw has a flange about ¼ inch behind the screw head—the bracket clips into the space between flange and head.) There is no exit route for the speaker cable when the speaker is mounted tight against the wall—no notch or groove, that is, running out of the recessed terminal block. The setup directions recommend cutting a hole in the wall just behind the speaker terminals and running the cable down between the studs. This is certainly the most tidy arrangement, but isn't an option for many potential buyers.

For my setup, I had a wood shop cut particleboard panels the same size as the back of the S30, with notches in the panels providing space for the speaker cable to pass between panel and speaker. I mounted these panels to the wall, screwed the shoulder screws to the panel using the provided template (making certain that the terminal block would be lined up with the notch), connected the cable to the speaker, and finally mounted the speaker to the panel while carefully routing the cable through the notch. Another option would be to loosen the shoulder screws enough to allow the cable to pass behind the speaker, but this would result in a decidedly more wobbly setup.

The B15 subwoofer was positioned in the right front corner of the room, with the back facing out to provide easy access to the equalizer controls. Corner placement for subwoofers is a little controversial, but the B15's owner's manual states that “placing a subwoofer in the corner of a room will result in the maximum number of ‘peaks' and the minimum number of ‘dips' in the response. Most dips cannot be equalized . . . "

Once the subwoofer is positioned, however, how do you dial in those equalizers? Revel recommends test equipment with a resolution of at least 0.1 octave down to 20Hz, which leaves out most commercial real-time analyzers. For my setup, Kevin Voecks visited with his computer-based test equipment and measurement microphones. Because you're unlikely to have the benefit of such gear, Revel has provided what they call their Low Frequency Optimizer (LFO). This consists of a test CD, a computer program, and a sound-pressure-level (SPL) meter. The buyer must provide the SPL meter; the low-cost RadioShack models are suitable (analog version recommended). Revel measured a large sample of these meters and found that although their response was not particularly accurate, the deviations were consistent from sample to sample. That is, all the units measured largely the same, so the deviation from flat response could be compensated for by the LFO program. (For more on using Revel's LFO, see the accompanying sidebar.)