Bang & Olufsen BeoLab 5 Speaker Page 2
My colleagues and I were therefore astonished when the curtains parted and the BeoLab 5 was revealed in all its futuristic glory - nothing in its extraordinary appearance immediately explained the sound we were hearing! From the explanatory technical seminars that followed, we learned that the 38-inch-tall speaker does not even employ any radically new or exotic drivers, just a 3/4-inch tweeter, a 3-inch midrange driver, a 6 1/2-inch midbass driver, and a big and heavy 15-inch woofer.
I admire B&O's Danish reserve in not calling the last a "built-in powered subwoofer." Perhaps the reticence is because the entire speaker is powered - there's a separate digital (Class D) power amp for each driver. B&O rates them at 250 watts for each of the higher-frequency drivers and a whopping 1,000 watts apiece for the midbass and woofer. With a total of 2,500 watts on call, no wonder the BeoLab 5 can play so loud and clean.
You don't need to buy any power amps for a system equipped with BeoLab 5s. In fact, to get rolling, all you need is a DVD or CD player with a standard coaxial digital audio output, which the speaker feeds into a digital signal processing (DSP) chip. The supplied universal remote lets you turn the speakers on and off, adjust volume, and control the player as well as other gear. The speaker accepts line-level analog audio signals, too, but they go through analog-to-digital conversion so the chip can work its wonders. And wonders they are.
The DSP engine is an electronic crossover as well as an equalizer. Digital processing is also used to monitor the heating of the drivers' voice coils. If they're ever in danger of overheating, the system will shut itself off. But it's what the DSP chip does at the low frequencies that constitutes one of two outstanding innovations of the BeoLab 5 - a process called Adaptive Bass Control. Simply put, digital equalization is applied to the midbass driver and woofer to correct for the deleterious effects of room acoustics - the resonances and "suckouts" that can make bass sound very uneven.
You activate the Adaptive Bass Control's self-calibration system by touching the center of the brushed-aluminum top plate for a few seconds (calibration is done separately for each speaker). The speaker starts emitting a series of moderately loud low-frequency test tones. These are picked up by a built-in microphone mounted on a movable rail that runs underneath the bottom-mounted, downward-radiating woofer. The speaker measures its own "close-miked" response, similar to the way we measure bass response in our lab tests. Then the response-measurement process starts over, but - and this is the neat trick - with the microphone extended a little further out into the room (see photo above).
The differences in the two responses are always minuscule, but there are differences, and they vary with the speaker's placement in the room. They are caused by changes in the "radiation impedance" the speaker "sees" at different frequencies. Think of a soundwave from a room reflection trying to push the woofer cone in while the amplifier is trying to push it out - you have increased radiation impedance leading to a partial sound cancellation and a dip in the response further out in the room. It's these room-caused cancellations and reinforcements that the DSP chip measures and removes by introducing appropriate boosts and cuts in the signals feeding the low-frequency drivers.
This system cannot fully correct the bass response at the listening position. Instead, it corrects the output of the speaker wherever you put it, which is different though closely related. But in our listening room, the sonic effect was at times enormous and always an improvement over the noncalibrated response. With one speaker placement, the peak-to-dip span of the low-frequency response curve at the listening position before calibration was some 20 dB, which is typical for a normal listening room. This was reduced after calibration to a span of only 11 dB, a considerable improvement that was clearly audible as tighter-sounding, less boomy bass.
The second major innovation of the BeoLab 5 is responsible both for the strange appearance of the top half of the speaker and for its excellent imaging. As you can see from my closeup of the midrange, that driver points upward and into a reflector, as does the tweeter. B&O licences this arrangement, called Acoustic Lens Technology, from Sausalito Audio Works. The midrange and tweeter are each at one of the two focus points of the ellipse-shaped reflection surfaces, which serve to make their radiation appear to come from a 180° horizontal arc. This explains both the very wide horizontal dispersion of the speaker (click to read "tech notes") and its sit-anywhere imaging qualities.
The saucer shapes, aside from giving the speaker a distinctly Star Trek appearance, also help reduce its downward radiation and the effects of the typical floor bounce on imaging and timbre. The lowest saucer also has a finger grip molded into its underside, which is useful when moving the speaker (it weighs about 134 pounds).
Listening to the BeoLab 5 back home in Sound & Vision's New York City sound lab more or less confirmed my overseas impressions. But listening with familiar source material - like Lorin Maazel's two-microphone recording of Respighi's orchestral showpiece Roman Festivals (Sony Classical) - enabled me to pinpoint the speaker's strengths and weaknesses. It can play very loud, very clean, it has a very smooth tonal quality free of major colorations and resonances - voices came out particularly well - and it can produce good stereo imaging from a very wide listening area and extremely good imaging when heard from a prime listening position. The slight swayback shape discernible above 1 kHz in our graph probably caused the slight lack of brilliance I heard on orchestral strings as well as in the cymbals on many well-recorded jazz and pop CDs. But these effects are subtle. B&O's BeoLab 5 is the best speaker I've heard in some time, and it embodies extremely important technical innovations that deserve further exploitation.
Given that just one speaker is priced at $8,000, I can't imagine B&O trying to market a still costlier version, but this could be the start of something smaller. In Denmark, I heard some BeoLab 5 derivatives mounted on the dashboard as the front left/right speakers in a car system. The system needed some refinement - remember what a big to-do it was on Star Trek when they separated the saucer section of the Enterprise! - but on the basis of the marvelous BeoLab 5, I have no doubt now that B&O engineering will eventually make it work. I'm looking forward to a sequel.