Index to Articles in this Issue Home Page

 

Product Review - Ambiance Acoustics California Cube Loudspeakers - April, 1998

Karl Suager

Divider

 

calcubeenclosures.jpg (6 kB) Ambiance Acoustics Cal Cubes

Four 4 1/2" Drivers in Each Enclosure

Mfr. FR: 38 Hz - 16.5 kHz -5 dB

Sensitivity: 90 dB/w/m

Power Rating: 10 w min - 250 w max

Size: 13 5/8"H x 13 5/8"W x 13 5/8"D

Weight: 30 Pounds Each

Price: $1,395 - Two Speakers and EQ

 

 

Ambiance Acoustics, P.O. Box 27115, San Diego, California 92198; Phone 619-485-7514; Fax 619-675-9245; E-Mail sales@calcube.com; Web http://www.calcube.com

Yep, they’re cubes all right. These particular cubes sent by Ambiance Acoustics were finished in what they call "Rust", but looked more to me like dark, iron-rich granite. Maybe "Dark, Iron-Rich Granite" would confuse a potential customer, so they just went with "Rust."  The finish itself was kind of attractive, a nice change from oak, cherry, and the various shades of black passed off as other than the ultimate in dark. The six surfaces didn’t meet at the seams as well as they would have if each were cut at 45 degrees and butted against each other. Rather, the laminate laid on itself like a T that had been pushed to an L. This technique probably proves much more cost effective, but does expose the cut side of the laminate, making it apparent that the boxes aren’t the solid material that the surface might otherwise lead you to believe. Still, I didn’t notice it from a few feet away, and they are kind of chic, like I’d expect to see on Beverly Hills, 90210.

The cubes carry four identical 4 1/2" drivers, mounted inset behind a detachable grille fastened via Velcro. The manual says that they sound better with the grilles off. I certainly think that, like most speakers, they look better that way, so off they go! Even though the pair we received had a split seam due to a ding acquired in transit, they’re otherwise built fairly well. The cabinets weren’t a match for the exceptionally damped (via Tekna Sonic) NEAR 15Ms reviewed awhile back, but few are, and I hardly consider it a fair expectation. The drivers weren’t exactly flush with the cabinet or uniform in angle or depth either. I don’t think that it’s really a big deal, other than cosmetically speaking, as flush mounting tends to be most important for tweeters since they emit the highest frequencies that are most prone to diffraction artifacts. With 4.5" drivers, the treble will beam (like a laser) so that very little, if anything, will roll off the edge of driver, and smear off the baffle. Still, though, I would have appreciated an effort towards perfection.

At the rear, underneath a port that looks like the exhaust for a small jet, the gold-plated, five-way binding posts are standard moderate high-end issue, capable of a good connection with any banana plug, spade lug, or heavy gauge bare wire to boot. I like bananas since they allow quick hookups and a decent connection, important when you want to swap speakers before the aging mind forgets.

The port itself spans a full 4 inches in diameter. JEJ asked if it really moved that much air, and I did think it a little weird. Four of these drivers have the same surface area as a single eight inch woofer, but widening a port by itself tunes the box up, contrary to what one might intuitively assume. Since the port on these speakers is only 5 " deep, the dimensions of the box would yield a tuning frequency of about 65 Hz.

The EQ (PC board shown in photo at right) includes a "subsonic" filter at about 60 Hz. Actually, this is a genuinely useful feature for the dynamically inclined. With ported enclosures, there is little if any output below the tuned frequency anyway, and the woofer can easily unload below that point, contributing little to the sound pressure of the room since at that point the contributions of the port and the woofers tend to cancel. Cal Cube Equalizer PCB (3943 bytes)The electronic filter, before amplification, saves the amplifier the hassle of deep bass, and minimizes driver distortion by lowering the excursion demands, keeping the coil in the most linear range of the magnet assembly. I wouldn’t be too worried about protecting the drivers themselves. I stupidly punched in the first EQ from bypass directly into my Aragon, a direct-coupled amp, and before the DC sensing relay could disconnect the outputs, the 8008BB had fried a 10 amp rail fuse, equivalent to about 600 watts into the respective load, for a good second or two. I wasn't happy about it, but after receiving a new EQ, and obtaining new fuses, everybody was ready to play nice again.

Design Notes

What we have here is a multi-driver, somewhat full-range, vented system with active equalization. EQ in consumer audio catches a lot of flack. Many people quote the inherent phase shift of filters. What they may not realize is that the drivers also act as an acoustic filter, so that a driver with a flat response due to EQ, will have an end phase response better than a driver without active EQ and a not so flat response. The Cal Cubes not only benefit from active EQ, they NEED it. Without it, they sound just awful - muffled, boomy, and all out gross (probably the same as most speakers would sound without their crossover network).

Designing a true full-range system might seem simple with the use of an EQ to boost the top and bottom but, when you get into the specific physics of dispersion patterns, power response, diaphragm resonance, and bandwidth limitations, it’s so very far from simple. That explains why most loudspeaker companies use drivers specialized for different frequency ranges (woofers, mids, and tweets.)

Full-range systems do have some advantages. For instance, choosing to run a driver (or drivers) full-range eliminates the need for a passive crossover, as well as the challenge of blending the response of different drivers together. In addition, there are no crossovers to cause phase shifts. I’m not entirely convinced about the significance of phase shift effects by themselves in loudspeakers when it comes to audible differences, assuming a competent design. It’s well known and accepted, though, that phase shifts between separate loudspeakers can be very audible, as it is fundamental to our sense of direction. However, over a narrow band, and from the same direction, it has been shown by researchers in the field that the audibility of phase shifts by themselves have been minimal. That’s not to say that a poorly designed crossover and driver array can't induce phase differences which cause peaks and dips in the tonal response, or that a poorly matched set of crosover components and/or drivers can't cause differing phase shifts between speakers and smear localization. In any case, how does this apply to the cubes? By not having a crossover, it simply avoids a design problem which could otherwise hamper performance.

It is important, though, to note that the lack of a crossover does not make the speaker immune from phase shifts inherent to the drivers themselves. On one end, all drivers have a bottom resonance, a frequency at which they store energy and dissipate it depending on their own damping properties. At this resonance, the driver releases energy into the environment later than frequencies above or below that point, causing what are called group delays. How long the delay is depends on the damping (energy absorption) of the system. Without getting into electrical and mechanical properties that affect damping, know that a sealed box damped for the flattest in-band response induces the same phase shift as a crossover with the same frequency response, or a 2nd order, 12 dB/octave slope.

A ported design uses the moving mass of air through a tuned port, or perhaps a passive radiator, as a resonant device which will absorb energy from (and hence limit the motion of) the active driver and thereby extend the lower end response of the system and lower distortion at and slightly above the tuned frequency. But, because it is another resonant system, it also induces a secondary delay to that of the original driver. It’s not a hard and fast rule that sealed systems have better phase response than ported systems, since the interaction of the two resonant systems is more complex than simple summation, and the damping of drivers designed for ported use tends to be higher due to system requirements. But, it would be a false statement to say that these smaller full-range drivers are any faster at 60 Hz than a well-damped 12 inch woofer in a sealed enclosure, or even a well-done 12 inch woofer in a ported enclosure which reaches a 30 Hz cutoff. In fact, the opposite would be true.

So the possible advantages of this design aren’t in absolute phase performance, but that there’s minimal phase shift in the more critical band between 500 Hz and 10 kHz. How minimal depends where you’re listening. I say this because with multiple drivers, each driver has its own arrival time, and any offset in distance results in a phase shift. A few degrees in lateral movement can result in significant phase shifts in the upper band. For instance, 15 degrees will offset the drivers by about 1.5 inches, which at 4 kHz results in a 180 degree phase shift (half the wavelength of a 4 kHz signal). That, in itself, isn’t such a big deal, since it’s the phase shift associated between drivers of a 12 dB/octave crossover, which is relatively light. However, if not dealt with in terms of the response at the listening position, causes lobing (off-axis dips in response due to uneven dispersion of the loudspeaker) that can obscure musical information.

This, in itself is the biggest weakness of these loudspeakers, which is somewhat ironic, considering that controlling dispersion was an intentional design goal. Many speakers limit vertical dispersion to control unwanted reflections. THX speakers often do it by stacking tweeters. D’Appolito designs do the same thing with the midrange drivers, minimizing reflections from the ceiling and floor. The Cal Cubes intentionally limit dispersion through lobing in both the vertical and horizontal plane, beaming to the extent that if you straighten up, slouch, lay back, or lean forward, you lose half of what’s there. I think that they did it too well, because that's what really did them in for me, since I’m normally not able to enjoy music with my head locked into a vise.

Now, one could redesign the speaker, using the same driver array, to minimize this behavior. If a single driver ran full-range, and the other drivers gradually filled in, one by one, starting at about 500 Hz, it would reinforce the output in the bass where needed most to lower distortion, and keep the dispersion more consistent. But, then you lose the controlled dispersion which was the original intent, and it would require the use of a crossover. It wouldn’t be a difficult crossover to design. Just three inductors (one for each of the three extra drivers) and the EQ could be redone to accommodate the new frequency response. The phase shift wouldn’t be much, about 45 degrees at the crossover points. Or, if implemented in a completely active fashion with DSP algorithms, this could eliminate any phase shift caused by a crossover. However, it would destroy the whole marketing position that crossovers are bad.

That’s not to say that the Cal Cubes as is carry no merit badges. Though they don’t go very deep, and the transient response resembled the Polk RT-800s, punchy yet a little thumpy, the bass output can easily cause a few neighbors a satisfactory loss of gruntledness at higher levels without strain. The midrange flows easily, and the treble, while lacking any sense of shimmer, and sacrificing a bit on clarity, is far from fatiguing, possibly a benefit of the treated paper cones.   (Paper may not be all that stiff, but it doesn't screech either). All, in all, if you can set yourself in exactly the right spot, with good positioning of the loudspeakers, they might suggest what Bose 901s would sound like if you could actually derive some focus to the soundstage.  When the bodies align, they're a punchy, pleasant-sounding speaker that can play pretty loud.

They’re an interesting product, and bound to fit somebody’s palette. If you get the opportunity, check 'em out.

Karl Suager

 

Associated Components used for the review:

Infinity Renaissance 90 Loudspeakers
Aragon 8008BB Power Amplifier
M&K S-85, S-80, & LCR-55 Satellite Speakers
M&K V-75MKII Subwoofer
Passive Controller w/50 k Nobel Pot
DH Labs Silver Sonic Interconnects & Speaker Cable
Bybee/Curl Prototype Power Purifiers
API Power Pack V AC Line Conditioner
JVC XLZ1050 CD Player
Denon AVR-3200 Receiver

 

Manufacturer Response:

We'd like to thank you and Mr. Suager for taking the time and resources to review our California Cube loudspeaker system, which would fit well in BH 90210. Regrettably, Mr. Suager did not seem as enthused about our system as our customers are, but everyone is certainly entitled to their opinion. And when it comes to loudspeakers, especially those out of the mainstream, there are many opinions. Mr. Suager raised some points in his review that we'd like to comment on in chronological order.

A) The speakers Mr. Suager reviewed were demonstration models. Unfortunately, due to mishandling in shipping and the fact that these speakers spend more time on the road than in the shop, they were not in the "perfect" condition that is normally expected when purchased. This particular pair had a ding in the bottom front panel corner that when placed on stands, was unobservable. The argument can be made that for product reviews, a manufacturer would want to send its best pair. Well, the minute a speaker is packed up and shipped, it's a used item. If we sent out a new pair for every demo or review, we'd have an awful lot of used inventory, and that's not cost effective. We had also mentioned prior to the review that this pair were demos and not aesthetically reflective of what the customer receives. With respect to the comments about finish quality, the laminate we use is an eccentric imported Italian one that requires sensitive handling when the enclosures are manufactured. The laminate is applied after the box is built, so it's impossible to create a seamless edge. It would be cost advantageous to apply the laminate before cutting the box panels, which by the way are cut at 45s, but any mishandling while assembling the box may result in splintered edges. These are hand assembled. Some of our veneer enclosures use a much more pliable material that's applied before cutting, and is actually "rolled around" the side panels yielding no seams.

B) All cuts made to the enclosure are done on CNC including the driver recesses. These are exact cuts. The four drivers when mounted are equidistant from one another in a horizontal and vertical plane. The drivers have plastic diffraction rings that when torqued (in a cross-torqued method) may not seem perfectly aligned, but the driver frames, and subsequently the voice coils are in alignment. And as Mr. Suager correctly mentioned, edge diffraction is more consequential to high frequencies, and cone drivers being directional make the diffraction rings in this instance inconsequential to the sound.

C) The rear "exhaust for a small jet" port is large for two reasons: 1) elimination of port noise 2) on any square flat panel that is reinforced at the edges, the resonances are greatest at the center of the panel. The rear panel receives the primary rear wave from the drivers. If a large port is centered on this panel, you've effectively damped the rear panel as well as reduce standing waves. The front and rear panels are 1" MDF, with the sides 0.75" MDF.

D) Mr. Suager's dissertation on group delay and phase response contains some very positive points relative to the design of the Cubes. We do however take exception to the allusion of transient response in the Cubes vs. a conventional 12" sealed or ported system. A "faster" speaker system cannot be determined by porting alone. Variables such as driver force factor, suspension compliance, total Q, moving mass, air load, crossover type, etc., all enter into the equation. In energy time curve measurements which measure transient response coherency, there are very few if any, loudspeakers that can compare to the Cubes. We realize the Cubes are not flat to 20 Hz, but for a bookshelf size speaker that has usable bass between 30 Hz and 40 Hz, another feat, having this sort of time coherency is reflective of a well engineered design. As Mr. Suager correctly stated, it would have been easier to employ conventional loudspeaker technology, but we didn't feel it offered the benefits of our current design.

E) Mr. Suager's main complaint, a small sweet spot. Quite simply, a smaller sweet spot increases imaging by decreasing room interaction. While his suggestion about redesigning the Cubes with a progressively ramped crossover design, using only one driver for full range above 500 Hz does have merit with respect to off axis comb filtering, you'd still have a small sweet spot due to the inherent design of cone drivers. The two problems besides phase shift that are introduced by implementing such a crossover are more circuitry (distortion), and an imbalance of power handling between the four matched drivers. Mention of DSP correction was made to offset any nasties introduced by crossover circuitry, and although DSP is a marvelous tool, we believe a re-engineering of the Cubes is extreme.

With respect to the correllation between the Cubes and the 901s, we won't even touch that.

In closing, we've been around long enough to know that it's impossible to please everyone when it comes to loudspeakers. And unfortunately, Mr. Suager's preferences are different than what the Cubes offer. But we'd have been interested to know more about setup, program material (music or movies), and perhaps a bit more subjective critique of the overall sound. We do however appreciate Mr. Suager's time, and the bandwidth extended to Ambiance Acoustics for this review. Thank you gentlemen.

Sincerely,

Robert J. Salvi, President

Ambiance Acoustics

 


Copyright 1998 Secrets of Home Theater & High Fidelity
Return to Table of Contents for this Issue.