Celestion SL700 loudspeaker:
Page 2 The loudspeakers were carefully positioned for optimum performance, and the stands were all coupled to the tile floor beneath the rug with spikes. In addition to a rigorous listening test, with no other speakers in the room, the Celestions were used for everyday listening.

As was the case with my last loudspeaker report, this review took place while I was moving house. The listening sessions had to be broken into two distinct periods. The first took place in my old room, with whose acoustics and idiosyncrasies I am, of course, very familiar. The second set of listening sessions took place in my new room—a forthcoming "Matter of Taste" will include details of how and where the two rooms differ.

Celestion SL700
As I mentioned earlier, the Celestion SL600 has been my reference box loudspeaker for five years now, though this is not to say that I haven't been looking for a replacement. In fact, when the SL6S was announced in late 1985, with its aluminum-dome tweeter replacing the copper dome, and a dual-surround for the woofer to increase the intrinsic dynamic range, I thought that a replacement for the '600 couldn't be far behind. I was wrong, of course. It wasn't until the summer of 1987 that the '700 was announced, and not until earlier this year that production samples became widely available.

Although ostensibly the '700 is very similar to the '600, in that it uses an identically dimensioned cabinet fabricated from Ciba-Geigy's Aerolam honeycomb material, with a nominally identical drive-unit complement—metal-dome tweeter and 6½" Kobex-coned woofer—it is actually a very different animal.

But first, what is special about the cabinet construction? What's wrong with God's own wood? Though in general it is reasonably well-damped, wood in the form of panels does ring—consider the sounding board of a grand piano, for example. And the panels that comprise a typical speaker cabinet don't behave any differently. They have their own preferred frequencies where they ring like a (dull) bell. A traditional way of dealing with these cabinet-wall resonances has been to add mass and damping to the usual fiberboard (MDF) panels to reduce the amplitude of the resonances and move them down in frequency, away from the critical midrange. But as this also lowers the Q of the resonances, they are more likely to be excited, and the overall sound may become worse.

Furthermore, by adding mass, you increase the energy storage problems—you are increasing the size of the mechanical "capacitor"—and the time performance of the cabinet, the "smearing" of musical transients, could also worsen. It seems that this approach only works when the cabinet is sufficiently small that the fundamental resonances are high enough in frequency to stay out of the critical region when modified by mass-damping: the Acoustic Energy AE1, for example. Or when the amount of mass added is extreme enough, and well-damped enough, that the cabinet becomes truly dead, as in the case of the Wilson WATT.

The approach chosen by Graham Bank for the original SL600 was to replace fiberboard with a much lighter substance. For the same stiffness, panel resonances are pushed up in frequency, to 1.2kHz in the SL600, over an octave above the subjectively crucial midrange. Because there is less mass involved, there is much less stored energy able to be released after the event, the result being a "dead" cabinet, with very little sonic signature. However, as the walls are more transparent than usual to the internal air-space resonances, much more care has to be taken to damp these effectively. In addition, the '700 enclosure features the figure-8 bracing seen in the '6S, which should push panel resonances even higher in frequency, and is painted in a gray, textured, Nextel finish.

The result should be a non-cabinet, where the sound you hear is the true sound of the drive-units alone. As implied earlier, the '700 uses a version of the aluminum-dome unit first seen in the SL6S. Whereas the '600's copper dome has a breakup mode fairly low in frequency, the '700 dome doesn't go off until above 24kHz, doing away with the need for an individually tuned notch filter. In addition, the "mellow" treble balance of the '600 was, in my opinion, largely due to the fact that the designers just couldn't squeeze any more sensitivity out of the tweeter. The top octave may have been depressed, but it was all you were going to get from the relatively heavy dome. The aluminum dome, being lighter, gives more HF level with the same amount of drive, resulting in a more balanced treble.

The tweeter crosses over below 3kHz to a 6½"-cone, cast-chassis woofer, designed by Graham Bank and developed with the aid of laser interferometry to allow the woofer to act as a pure piston over its working range. At low frequencies, the entire cone acts as a piston; as the frequency increases, then the radiating area inevitably decreases, but instead of the rest of the cone going into uncontrolled breakup, which is often the case with paper cones, it supports traveling waves in the cone material, which do not transmit energy to the air and which can be absorbed by careful choice of surround material. This drive-unit actually has a two-piece surround, the inner concentric ring of which is of PVC, to correctly terminate the midrange traveling waves in the cone. The outer ring is made of soft rubber, to give a freer action at lower frequencies, resulting, it is said, in better articulation. Both drivers are manufactured by Celestion.