Snell Type A loudspeaker
Although you'd never know it from the boxlike external appearance, the enclosure design is quite unconventional. The front panel which mounts the mid- and upper-range drivers is curved like one side of a cylinder (fig.1). This eliminates the typical, abrupt corners at the edges of the front panels, thus minimizing the diffraction effects which cause sharp mid-range response irregularities and, hence, impaired stereo imaging. The system comprises two physically separate enclosures one for the woofer and the other for the two upper-range drivers, which stack one upon the other. Locating pegs between them keep them aligned. The two boxes are electrically connected by a short length of cable with dual banana plugs at both ends, so it is a lead-pipe cinch to separate them for bi-amplifying.
Fig.1 Snell Type A, the upper-range drive-unit enclosure.
Although we normally recommend bi-amping any speaker for best possible result it will not buy you as much improvement in the Snells as in most, because it is impossible to bypass the internal crossover element which rolls off the low end of the mid-range speaker. Additional attenuation there, from an external electronic crossover, will cause excessively sharp rolloff and, consequently, incorrect phase-shift relationships through the crossover region. The result will be a hole in the response there. The electronic crossover should be used only to attenuate the upper end of the woofer.
The upper-range amplifier should be operated full-range or, if there is such a provision on the crossover, rolled off below 70Hz or lower. Our tests were conducted with the speakers mono-amped; bi-amping will slightly improve all aspects of low-end performance but will afford little improvement through the upper ranges.
A two-position rear-panel switch on each speaker affects the system's high-end countouring, not by changing tweeter level but by changing the slope of the entire high-end response. This was one of the more useful controls of its kind we've encountered, for—whether by design or happenstance—it almost exactly compensated for the difference in high end between good tube and good solid-state amplifiers. With the switch in its "normal" position, we found the high end to be natural with tubes and a little hot with transistors; with it in the "decrease" position, highs were correct with transistors and dull with tubes. We used the appropriate settings for our tests.
How, then, did the As sound to us? On the positive side, they image superbly—as well as any system we have heard—and reproduce instruments in their proper apparent size. Their treble performance is excellent—extremely smooth and velvety but subtly closed-in at the extreme top. Overall balance is very good but the system has a somewhat warm sound rather than being entirely neutral. Bass is clean, subjectively flat to 30Hz (with transistors), and with very good detail (although not the equal of the better transmission line-loaded systems). But...
The system has a subtly recessed quality, as though the middle range is somewhat depressed (fig.2). The sound in this respect is similar to that of the Rogers/BBC LS-3/5a minis reviewed here recently, although the Snells have less of the midrange suckout than do the BBCs. But the fact that the Snells don't have the slightly whooped-up bottom of the BBCs makes amplifier mating rather more of a problem.
Fig.2 Snell Type A, subjective response with transistors (red) and tubes (blue).
Neither speaker really comes to life without the addition of a bit of brightness in the signal. With the BBCs, this was easily done by using a tubed power amp; the typical tube brightness offset the midrange depression, while the slight sparseness of the tubed amps at the extreme low end had no effect on the BBCs. They don't extend that far down. With the Snells, though, the added presence of the tubed amplifiers was offset by their tendency to make the low end somewhat loose and, at the same time, thin at the bottom. Solid-state amps, most of which elicited stunning low end from the Snells, did nothing for that slightly deadish midrange.
In short, no amplifier that we tried could yield equally good performance from the Snell through both the middle range and the low end, and we cannot think offhand of any amplifier that might fill the bill. As with the Acoustat Xs, a good equalizer solved the problem neatly and, in the case of Dynaco's SE-10, without adding any other signal degradation that anyone was able to hear (that is, until it was pointed out to some listeners that the equalizer was in circuit, after which they were able to hear "quite clearly" how much the SE-10 was muddying the sound). We are confident, however, that logic and common sense will not persuade any dedicated perfectionist to pollute his system with any equalizer, no matter how good.
The As will produce scads of volume without strain, up to well beyond the needs of most sane listeners, and they are fused in case you overdo things anyway. The only time we found ourselves popping fuses was as a result of our (defective?) Ampzilla II generating severe subsonic pulses at high power levels. But fuses are cheap.
For the audiophile, these are an ideal choice in their price range, as they do everything that he listens for quite a bit better than do many systems costing substantially more.
For the music listener who is intimately familiar with live-music timbres, though, we can recommend the Snells only if mated with another component in the system that will liven up the presence a bit, such as a tubed preamp (for brightness) and a solid-state power amp (for optimum bass performance).
All in all, though, these are really hard to fault without the kind of nitpicking we just succumbed to. Few people, we suspect, will be disappointed with them, which is something we cannot say of too many speakers in any price class.—J. Gordon Holt