Sound-Lab A-1 electrostatic loudspeaker Page 3
The main purpose of the panels is to enlarge the effective area of the speaker's baffle. As is well known, dipole radiators with a finite baffle size lose radiation output below a critical frequency due to front-to-back cancellation of sound energy. This critical frequency is determined by the average baffle dimension. Cancellation (at the rate of 6dB/octave) is effective for frequencies whose half-wavelength is larger than the average baffle dimension. At 50Hz, the half-wavelength is 11', so that dipole radiators must be large in size to maintain decent bass efficiency.
If the dipole is floorstanding, things are a bit better: the floor effectively extends the size of the baffle. The mirror image of the baffle in the ground plane actually increases the average baffle dimension by about 41%. Another way to increase the baffle size is to use wings, or "folded edges," as they were called in the hi-fi consoles of the '50s. Large wings produced a cavity resonance that gave reproduced sound a boxy quality. Sound-Lab insists that, with the panels flared out from the speaker frame, cavity resonances are unlikelyyou can have your cake and eat it too. The panels are said to increase low-frequency response, efficiency, and dynamic range, tighten up bass definition, provide a more up-front sound, and improve staging and imaging. Apparently, there are no tradeoffs with the wings in place.
In my experience with the wings, I found that they definitely extended bass response and improved bass dynamics in the process. It's really not fair to compare bass definition with and without wings. However, it is possible to mass-load the frames without necessarily using wings. To do this, I raided an old weightlifting set I found in the garage, left over from the days when I actually had time for physical fitness. I used two weights per speaker, hanging a 15 lb weight on one side and a 10 lb weight on the other. That was all I could scrounge at the time, so there's no reason to suppose that these are the ideal weights. But it worked. Midbass definition tightened up to the point that I didn't really miss this aspect of the wings.
Roger West had installed the A-1s in the Reference Room, wings and all. Then, however, came the slow process of optimizing speaker placement and room treatment to coax from the speakers their full sonic potential. Because the room is dedicated to listening, I had complete freedom in speaker placement. But at first I was reluctant to make major changes in the speakers' initial positions. Without wings, the A-1s are quite easy to move about, even on a carpeted floor, because of their casters. With the wings attached, it's a different story.
Initially I focused on getting the tonal balance nailed down. The first step was to experiment with distance relative to the wall behind the speakers, the listening seat, and the toe-in angle. In my 18.5' by 23' room, I ended up with the A-1s 6' from the front wall, 6.5' from each other (with only slight toe-in), and my listening seat only 9' from the front of each speaker.
Controlling the bass balance required locating ASC Tube Traps in all corners of the room. I used two Traps per corner: an 11"- atop a 16"-diameter Tube Trap, the Tubes' reflective sides facing outward. Other significant damping in the room included heavy drapes along the rear wall, a carpet over a thick foam pad, four 4' by 2' foam panels propped against the wall behind the speakers, and a small sofa. I was quite pleased with the resultant bass extension and control with the bass compensation set at the nominal position. High-frequency damping in the room was such that there was no problem in operating the Brilliance control fully openapparently a rare thing.
But the soundstage was like an LA freeway, the congestion and smog taking their toll on my patience. First of all, I couldn't see very far into the soundstage, certainly not to the back wall of the recording venue. And there was considerable broadening and blending of instrumental outlines, to the point of actively hindering the precise resolution of spatial detail. This was quite a mystery. I had felt confident that the back wave of the A-1, being in the dead end of a live-end/dead-end sonic environment, would be properly controlled. That did not turn out to be the case. I had to undertake another round of experimentation.
Originally, I had set up the A-1s nonsymmetrically in relation to the width of the room. The left speaker was about 2' farther from the side wall than was the right. In other words, the speakers' axis of symmetry was translated 2' to the right of the room's long axis. The folks at ASC felt that such a placement was a no-no, so I moved the speakers and wings to a fully symmetrical position. That helped some in stabilizing the width perspective of the soundstage, but little else.
Remember that the A-1s started life in my listening room with wings. I had yet to hear them without these albatross-like appendages; in a burst of energy, I removed them. Wow! I guess I wasn't ready for the resultant sonic transformation. The boxy, congested soundstage was liberated. It flowed and bloomed to fill the space between and behind the speakers in a flood of energy. Mind you, while spatial resolution did not improve, the overall spaciousness of the soundstage presentation increased dramatically. Unfortunately, bass extension without the wings was not as deep, and bass definition through the lower registers was not as tight. But I hated to give up the bass advantage of the wings; determined to give them one more try, I re-attached them.
About this time it occurred to me that my basic approach was flawedI had been relying entirely on sound-absorption treatment. To read any popular discussion on room treatment, you might get the impression that the ideal listening environment is an anechoic chamber: the more damping, the better. If I had a six-channel sound system to fool with, I'd be inclined to agree. With six channels I could reproduce ambient information from the sides and rear of the room in concert-hall fashion. The problem with two-channel audio, which we're stuck with for the time being, is that all ambient information is reproduced from the front. It's therefore important to involve the listening room in the reproduction process to enhance the reality of the illusion. That way, some of the original ambient information is reflected about the room and arrives at the listener's ears from the sides and behind the listening seat. This must be done carefully to avoid serious coloration. Some absorption is essential, but just as important is the use of diffusors to ensure that reflected energy is not specular (concentrated at a particular angle), but smoothly spread out spatially.
An important property of a plane wave is that it does not diffuse very much when reflecting off a wall; the planar nature of the wave front is pretty much preserved. It's important to actively enhance the diffusion process of a planar speaker's back wave. I would have loved to have tried some RPG Diffusors behind the A-1s, but I lacked immediate access to these animals. I tried the next best thing: the ASC Studio Traps reviewed by Jack English last February. Immediately behind each speaker I positioned a pair of Studio Traps raised to their maximum height and rotated so that the reflective side faced the speaker. That proved a real breakthrough. For the first time, spatial outlines snapped into focus to the point where I could sort out instrumental outlines with precision.
Another important benefit of this arrangement is that the back wave is controlled very close to the source. That way, the surface area that needs to be treated is minimized. Rather than covering a large section of the wall with diffusors and absorbers, only a pair of Traps sufficed close-in. Now there was no need to leave the four 2' by 4' foam panels against the front wall. I decided to move these; on a hunch, I tried propping a panel on the inside surface of each wing to provide further control of the back wave. That also helped define image outlinesnot to the extent that the Traps did, but every little bit helped.
That's how matters rested for a long while. Soundstage transparency had improved, as had spatial resolution, to the point where I was almost willing to live with the results. Maybe I'm just hard to please, but after a couple of months of this arrangement I was getting unhappy again. The soundstage wasn't integrating properly. It was subdivided, like a baseball diamond, into isolated left, center, and right fields. I still had to work too hard to resolve the depth perspective, and image outlines lacked the sort of spatial bloom that live instruments project. While imaging was okay, there was little magic to help the listener transcend mere hi-fi.
So I jettisoned the wings for the second and last time. As if Merlin had waved his wand, image outlines broke free of their straightjackets. The soundstage went into full bloom. Spatial outlines fleshed out in such a palpably 3-D manner that I was completely captivated by the magic of the illusion. Voices were so convincingly life-sized, so realistically sculpted, that I was sure I could just reach out and touch someone (footnote 2). As final tweaks, I positioned two tapered Traps against the front wall, with a RoomTune in between. Three additional RoomTunes behind the listening seat completed the room-treatment process. Rest assured that, at this point, I stopped fooling with the setup and settled down to some of the most enjoyable listening of my life.
At last the sound
I well remember my first exposure to an electrostatic loudspeakerthe classic Quad. The revelation that reproduced sound can convincingly mimic live music was based not so much on the Quad's inherent transient speed and lack of boxy colorations as on the "oneness" with which it spoke. I had become accustomed to the diverse personalities of the various drivers in a multi-way system. Even in a simple two-way system, there's considerable frequency overlap between tweeter and woofer. The upper range of a female vocalist would be reproduced by the tweeter, the lower registers by the woofer, and somewhere in the middle both the tweeter and woofer try to synchronize: two voices trying to reproduce one. Blending or integrating the drivers' personalities is one of the hardest tasks facing a speaker designer. Too often, the results lack complete coherence. With even a two-way of good design, I have little difficulty picking out the woofer's or tweeter's contribution to the overall sonic mix. The most successful multi-way dynamic speakers, in my opinion, are those in which the midrange is covered by a single driver.
Footnote 2: I should note that JGH, in whose ears we trust, has been using wings with his magazine-purchased pair of A-3s, and is pleased as punch with their performance in his room. He tells me that the imaging with the wings is better than ever. Imaging may be seventh or eighth on Gordon's list of priorities after tonal balance, timbral accuracy, dynamic range, harmonic textural purity, bass extension, resolution of inner detail, and Martin Colloms's pace and rhythm. Naturally, I'm dubious about his finding. But there's the outside possibility that the wings' effect on imaging is room-dependent, which means that they won't screw up the imaging in some rooms.Dick Olsher