Floor Loudspeaker Reviews

Sam Tellig has already written about the $1990/pair Thiel CS1.5 at some length in Stereophile, so you may want to look and see what he had to say about them (Vol.17 No.8, p.39). JA gave me the go signal to write up a full review with technical descriptions and everything, so don't blame me for redundancy. Just think of Sam's words as a Lead-Up instead of my words as a Follow-Up, and it all makes sense.

The CS1.5, Thiel's second smallest speaker, is a two-way system with a passive radiator mounted below the woofer. In Thiel's larger models there are plenty of drivers to handle the overlap in ranges dictated by their insistence on using first-order crossovers. In these the poor mid-woofer has to handle not only the bass and midrange, but also has to behave itself way up into the treble. At a 6dB/octave rolloff, the woofer's output will still be audible to 10kHz, a stratospherically high frequency for most 6.5" drivers. Can Thiel's extraordinary metal-cone woofer rise to the challenge, or will it fall flat on its faceplate like most average woofers?

The 6.5" woofer is quite a special aluminum-coned unit, with no dustcap, an exotic short-coil/long-gap magnet structure for lower distortion, and a cast magnesium basket. As if all of that weren't enough, the woofer also has a second, reverse-polarity magnet attached to the back plate to focus the total field even more uniformly into the gap. It also reduces the stray field so your neighborhood Gauss catcher won't impound the poor little guy, not to mention making it practical for placement close to a video monitor. There is also a copper sleeve on the magnet's pole piece to lower distortion even further. The 1" metal-dome tweeter has a large magnet and a vented pole.

Thiel has tough assignments for their crossovers. Here they expect it to not only correct the drivers' frequency responses, but also to synthesize a virtually perfect measured 6dB/octave slope in the crossover region, and to present a very smooth impedance, between 3 and 5 ohms at all frequencies. In order to accomplish this, the crossover has to be very complicated.

If you look at one of Thiel's cutaway views, you can see bunches of doodads all stuck to the inside of the speaker. Not to worry—that's not a nest of Borg insects. Thiel uses high-quality parts, even going to the trouble to bypass the series tweeter capacitor with a small-value polystyrene and tin-foil cap.

Still, I can't help but wonder if all those parts don't mess up the music a little bit. Maybe these speakers would sound even more transparent if they threw some of their "first-order perfection" theory out the window for the sake of more simplicity. For example, there's a notch filter in the crossover to smooth the woofer's diaphragm resonance at 7kHz, and this also restores linear phase behavior in that region. What if, instead, the crossover increased to second- or third-order in this region to try to eliminate all of the chaotic output from the stiff cone? Wouldn't no rough sound be better than phase-coherent less-rough sound? But what do I know? I'm a speaker user, not a designer.

Thiel goes to great lengths to ensure that their speakers will have a very linear phase response. That they actually publish a specification for it is unusual, and shows the importance they place on it. The synthesized first-order crossover, the shaping of the individual drivers' responses, and the physical arrangement of them on the baffle all allow such time-coherent behavior. Basically this is important because it means that all the different components of the musical waveform will arrive at your ears synchronized in time, as they would in real life.

The 1.5's cabinet is 1" thick, and well-braced on the inside. The front baffle is 2" thick in places, and sloped so that the speakers need little or no tilt to achieve the proper listening axis. The edges of the baffle are not curved like the other speakers'; Thiel decided to instead curve the edges of their grille frames to reduce diffraction. Their frequency response was optimized with the grilles in place, as that's how they're meant to be used.

I stripped a couple of the input binding posts (hexnut plastic-capped types) when changing speaker cables. I tried to tighten them and they just kept going round and round. I really do like to have cables held on by more than just the occasional lucky puff of wind. You manufacturers out there—you know who you are—let's see some case-hardened titanium binding posts that come with their own lug wrench and have 4"-long rotational braces inside the cabinet. Hopefully you won't get too many lawsuits from people claiming to have thrown their backs out while attempting to loosen the "Cable Drydocks." But seriously, the location of these posts is a pain even if you don't strip them. It's difficult to fit large spade-terminated cables into the space without bending them sharply, and it's virtually impossible to put them on unless you have the speakers tipped over precariously.

Sound
I started out with the CS1.5s in the same positions as the other speakers I review this month: positions fiendishly calculated to make the bass nice and smooth in the room. At 17" and 20", these positions were closer to the front wall than Thiel recommends; I felt like such a bad boy abusing their fine products so. Actually, I like the speakers so close to the front wall because I usually listen with the volume very quiet, and the bass balance of most recordings would otherwise be too light at such volumes.

They sounded good in these positions, and I left the grilles on so they'd have the proper balance. I found Thiel's toe-in recommendation spot on; ie, firing straight ahead toward the back wall, not toward the listening seat. They didn't seem to need any tilting; I preferred the sound with the bottoms parallel to the ground. If you have a taller listening seat than mine, though, and you prefer nearfield listening, you'll probably want to tilt the 1.5s to be on the optimum axis.

Detail, detail, detail. These speakers let me hear into Steve Tibbetts' recent masterpiece The Fall of Us All (ECM 1527) enough to hear its minor shortcomings. I could hear the slight loss of detail and the soft top octave resulting from Tibbetts' analog recording equipment. The Unitys did not reveal these small inaccuracies as well. In general, I think this album has very good recording quality; I can only find fault with it in comparison with super audiophile discs. All the exotic little metallic instruments and percussion sounded particularly lifelike on the Thiels. I think the Coherent Source thing they have going is really far out. There is a wholeness to the CS1.5's sound similar to what I hear from my B&W 804. It's not a trivial thing to arrange to have your sound groups all arriving with the proper time relationships, but I find it greatly improves the illusion of reality.

These Thiels are real thoroughbreds; they might even reveal recording-quality differences a little better than the 804s. Listening to Brian Eno's tripped-out space music on the Instrumental boxed set (Virgin/EG 39110 2), it all sounded perhaps a bit more detailed and involving than I was used to. This was good. Of course, calling this style of music "involving" might be a bit farfetched, but it seems to make a lot of sense late at night. The "High Definition Remastering, Super Bit Mapping" seen on this box is not just a cute little logo. This high caliber of digital mastering really delivered the sonic goods, and benefits from the super accuracy of a pair of Thiels. You'll believe that you're actually relaxing in Eno's home studio at two in the morning, surrounded by the soft colored lights from multiple keyboards and synthesizers, sipping Jasmine tea, and staring like a zombie into a big fishtank while contemplating what it would feel like to float around in outer space.

Unlike the Unitys, the 1.5s didn't have a bass balance that reminded me of minimonitors. They went down to 35–40Hz, and were neither boomy nor lightweight. The bass was smooth enough, and served the music rhythmically enough for me not to notice it as an individual entity very often. Instead, I was usually only aware of the bass as an integral part of the whole of the music. They did have slightly less bass overall than the B&W 804s, but this may be the more accurate balance.

The CS1.5 was also very smooth in the midrange. Instrumental timbres seemed quite accurate with good source material.

Listening once again to Journey's Greatest Hits, it was obvious that the speaker treble was neutrally balances. (Love those compilation albums for reviewing!) The recordings sounded definitely too bright. There was too much treble obscuring the midrange and yielding unnatural timbres. The 1.5s presented the recordings as they are; they were still enjoyable, but weren't given as natural-sounding a balance as they were by the Swans Batons. Of course, in a perfect world the recording engineer wouldn't have turned all the treble knobs up to 8 while rolling his eyes in their sockets and cackling evilly.