Apogee Caliper loudspeaker

John Atkinson, November, 1986

"I am not in love; but I'm open to persuasion," sings Joan Armatrading in her song "Love and Affection," the track I was playing when I finally realized that my attempts to get a sound from the Apogee Caliper ribbon speakers approaching what I had heard at the 1986 Chicago CES were bearing fruit. And that sentence pretty much describes the creed of the professional audio critic. Each new product that arrives at your door could be the one to pass the J. Gordon Holt "goose-bump" test, to leave the hairs on your arms permanently erect. Did the Caliper full-range ribbons excite my previously quiescent nerve-endings? Did Bobby Ewing return from the dead? Did Sam propose to Diane? Will Alan Alda ever outgrow Hawkeye? What on Earth made Georgette marry Ted Baxter? Why can't Tubbs roll up his jacket sleeves like Crockett? How could a fine actor like Jack Klugman accept such a dreadful role? Some of these questions will be answered overleaf, but in the meantime, what is a ribbon speaker?

The moving-ribbon principle for generating sound is as old as the loudspeaker, the prewar German Blatthaller PA system that announced the departure of Robert Mitchum by Lufthansa for all parts of Occupied Europe in The Winds of War (footnote 1) being a very large ribbon speaker. It is also one of the most simple. An alternating current is passed down a flat conductor suspended between the poles of a magnet; Fleming's Left-Hand Rule would indicate that the conductor should move back and forth at right angles to the directions of both magnetic field and current; and move it does, generating sound. It doesn't generate much sound unless the magnets are very powerful (read expensive) and/or the ribbon very large. The very low intrinsic impedance means that an average amplifier will be hard put to deliver sufficient current.

The ribbon does have advantages, though, over a conventional drive-unit. As with an electrostatic design, the diaphragm is driven over its entire area, giving an even dipole response above low frequencies, where the back wave will cancel that from the front, and the lack of cabinet implies the absence of a major source of coloration. Equally as important is the lack of coloration from diaphragm resonances. Apart from the one low-frequency resonance due to its suspension—and this will be very low if the ribbon is suspended floppily enough—there is hardly anything to color the sound.

In general, ribbons have been used only as tweeters (footnote 2): a small size is appropriate due to the need to reproduce only high frequencies, which leads to physically manageable (and affordable) magnets; the drive is via an impedance-matching transformer; and the low sensitivity is often compensated for with some kind of horn. The only large-scale manufacturer of a full-range loudspeaker resembling a ribbon is Magnepan, whose proprietary drive-unit—a grid of wires bonded to a Mylar diaphragm—acts in an analogous manner. Magnepan also uses a true ribbon tweeter in their MGIIIA and Tympani IV models. Nobody, however, was making a full-range ribbon loudspeaker, the theoretical advantages being outweighed by real-world disadvantages.

Such was the situation when Apogee Acoustics' designer, Leo Spiegel, retired from the aerospace industry at the beginning of the 1980s. Leo felt that a true full-range ribbon speaker would offer significant advances in clarity and neutrality over dynamic and electrostatic speakers, and with his son-in-law Jason Bloom duly set up a loudspeaker company to attempt the task of making such a speaker. Their first product, launched at the 1983 SCES, was a 7'-tall, full-range three-way ribbon loudspeaker, a system with astonishing dynamic range and bass extension, but, at over $6600/pair, hardly a volkspeaker.

Treble and midrange were handled by 80"-long, floppily suspended, corrugated aluminum ribbons hanging between the poles of a ceramic-magnet array in a steel frame. The bass unit, however, was a new departure in speaker design: a trapezoidal sheet of corrugated aluminum foil, reinforced with a transparent insulating Kapton tape on the rear, had a pattern of slits hand-cut in it, making, in effect, a large, flat coil. This coil is then mounted in front of an array of bar magnets, their poles so arranged that, as current flows through the aluminum foil, each section moves in the same direction. Unlike the tweeter/midrange ribbons, which are so floppy that they can hardly be said to be under tension, the woofer diaphragm is stretched across a wooden frame, much like a very thin metal drumskin. The art of the Apogee woofer lies in the fact that, coupled with the irregular shape, the tension is arranged so that, above the fundamental low-frequency "drumskin" resonance, the diaphragm will be effectively non-resonant.

The large number of magnets required and the labor-intensive production of the large diaphragms imply a size, weight, and price that rule out the possibility that the original Apogee would feature in any but a very small number of systems. Leo and Jason's game-plan, therefore, was to produce smaller and more affordable loudspeakers, which would sacrifice dynamic range, power handling, and bass extension but keep the key sonic attributes of the ribbon principle. Each year since 1983 has seen the introduction of a smaller Apogee ribbon speaker: first the three-way Scintilla at $3300, then the two-way Duetta at $2780, and now the Caliper, effectively a two-thirds size Duetta, costing a "mere" $1650.

The Caliper uses the smallest Apogee woofer yet, with an area of just 450 square inches (equivalent to the radiating area of three conventional 15" drivers!). The tweeter appears to be a cut-down version of the Duetta ribbon: three vertical strips of corrugated aluminum foil (the center one is slightly narrower), joined with a yellow Kapton tape backing, hang between the poles of the magnets. Closer examination, however, reveals that whereas the earlier Apogee tweeters hung free, the Caliper tweeter is clamped from the rear about two-thirds of its length from the bottom, and the top third "hinged" back slightly. This dogleg shape is said to improve the dispersion. Nearfield measurements indicated that the tweeter handles the range above 800Hz or so, with what appeared to be 12dB/octave slopes. Actually, the crossover is more sophisticated, gentle initial 6dB slopes giving a seamless transition region, with steeper slopes being used out-of-band.

Footnote 1: I think I've been watching too much television this summer; something to do with the NAD MR20 TV monitor I bought on BS's recommendation! But don't the Dr. Pepper ads have great sound? And you should hear how the Calipers handle the soundtrack of Miami Vice.—John Atkinson

Footnote 2: And as microphones. The superbly uncolored Coles/BBC 4038, used by Sheffield for their Firebird recording, is a ribbon.—John Atkinson



As with the other Apogees, the Caliper is available in two finishes: grey and taupe. Usefully, both driver/crossover connections are brought out on the rear to separate sets of five-way binding posts. The Caliper can be used with a single set of speaker cables, supplied jumpers paralleling the pairs of sockets; with two sets of cables, one amplifier driving the units individually (biwiring); or with two amplifiers, with each drive-unit connected to its own amplifier (biamping). Setting up proved to be no problem: the speaker panel is supported by a single angle piece at the rear; an adjustable foot tilts the speaker until a plumb-line (supplied) hung from the top matches a mark in the support; the listener is then on the right axis. (Apogee's instructions for unpacking and set-up are a model for other companies to follow.)

Taking the measure
I initially used the Calipers conventionally wired with Monster Powerline 2. Amplification was Audio Research SP-10 II and Krell KSA-50—Apogee does recommend 100W amplifiers to get the full dynamics—and the front end was Linn LP12/Ittok/Koetsu Red and California Audio Labs LP and CD players, respectively.

How was the sound? Well, as I indicated a few hundred words back, it was not easy getting the speakers to "sing" as they had at Chicago. After some experimentation with room position to get the best tonal balance between mid and upper bass, the balance was musical and the imagery was exceptional, particularly regarding depth. There was, however, a veiling in the lower midrange, with a thick upper bass and somewhat undercontrolled low bass.

With the other Apogee speakers, I have found biwiring to be mandatory. The low impedance, particularly of the Scintilla, means that cable "signatures" become exaggerated. The Caliper has a nominal impedance of 3 ohms, compared with the Scintilla's 1 ohm, so you would have thought that this would have been much less of a problem, but the sound I was getting did smack of "Ultra-Monster"; perhaps I should investigate biwiring. Remaining with the KSA-50, I hooked up MIT Music Hose on the bass panel, keeping the Powerline 2 on the tweeter.

Bravissimo (footnote 3). The James Boyk Chopin recording now had the character of the real thing, the left hand of the piano having weight and authority but not interfering with the natural presentation of the soundstage. The lower mids were still less transparent than I would have liked, male voices acquiring a little too much chest tone, but from a couple of hundred Hertz up the Calipers were just superb. Voices had a freedom from grain and a liquidly smooth harmonic structure, with no awkward breaks in tonal character around the crossover region.

The extreme highs were clean, but lacked a little "air." This neutral presentation of voice made enjoyable even such flawed recordings as the Columbia collection of optically-recorded soundtracks for the '30s Fred Astaire movies. The sonic scum is somehow separated from the music, allowing you to appreciate even more—if such a thing were possible—what a superb vocal craftsman old Twinkletoes was, even with his light, not particularly "beautiful," voice.

And when I put on Frank Sinatra's Capitol recordings from the '50s...They knew how to record big bands in those bygone days; maybe the acoustic bass should stage a comeback.

The surprising thing is that the Caliper can produce low bass at all, given the size of the diaphragm. Although we are not talking stygian depths here, such contrabass salsa as the octave doubling of the bass line in the Red & Blue Mix of Yes's "Owner of a Lonely Heart" came over at the correct level, without overload, distortion, or any one-note effect. Impressive indeed.

The area of soundstage presentation is very important to me; I find deficiencies in this area more annoying than tonal balance anomalies (if the latter are not too severe). Unless the producer has used such sonic necromancy as Aural Excitation, or other such deliberately induced phase anomalies, instrumental and vocal images should be small, and placed within their appropriate shroud of reverberation. If that reverb and ambience is coherent, as it is on a naturally miked classical recording, or as it can be on a rock production, then those small images will be set back behind the speakers, the exact depth depending on the direct/reverberant sound ratio of each. With speakers unable to preserve these spatial relationships, the soundstage turns to garbage, the ambience now bearing little correlation with the soundsources, and the image withdrawing into a flat plane between the loudspeakers.

A favorite test track of mine in this respect is "New Grange" from the classic Clannad album on RCA. This track starts with a distant reverberant thrumming which draws closer and closer, leading to a tasteful mix of guitars and drums behind the female singer during the verse; at the chorus, when the male voices add just the right amount of Gaelic, the soundstage expands to encompass the listener—except that on most loudspeakers it just doesn't happen; the sound only gets louder. With the Calipers, as with Scintillas, the image gets "bigger" as it gets louder, with a sense of freedom from restraint. Adjectives like "majestic" come to mind. And with the "Love & Affection" track, the image depth extended from the tip of my nose to the mountains east of Santa Fe.

You might think I exaggerate, but let me tell you a tale: one Larry Archibald has often written in these pages of his dissatisfaction with the CD medium. This is not due to any prejudice—the music just doesn't hold his attention. He starts to think about publishing, or business, or working on Mercedes automobiles. Take the Calipers biwired from the KSA-50, place the live Harnoncourt Messiah recording on Teldec in the tray of the Tempest CD player, and then watch LA smile. Depth, space, but above all, music to delight even his jaded digital palate!

Footnote 3: I'll say. Having been present for a brief "before" and an extended "after," I can only say that the difference between single wiring and biwiring was mind-boggling. I've heard A-B comparisons between different speakers that were much less dramatic. In the single-wired mode (at least with Powerline 2) I would find it difficult to recommend the Calipers at all.—Larry Archibald



It's not all cordon bleu catering, though. In addition to the potential for veiling in the lower mids, the low bass appears continually to be on the verge of being overcooked. This is not a problem with most recordings, as there is seldom sufficient subterranean energy to lead the panels astray. Occasionally, however, a recording happens along that is ideally suited for provoking misbehavior. The Reference Recordings Church Windows LP, for example. With the speakers driven by two Audio Research D-250s (see later), Side Two is cruising along in the mid-90s, spl-wise, without sounding too loud. Respighi then adds his deep organ pedal notes: the diaphragms attempt to go down through the gears to cope with increased load but rattle against the endstops around 102dB (flat, peak) (footnote 4), and one is left with the sense of having witnessed something distasteful. Or the ridiculously overdone bass note at the end of Tina Turner's "We Don't Need Another Hero" 12-incher. Peaking at a measured 104dB in midbass caused all manner of rattles. Such lack of control makes an Englishman feel uncomfortable.

At the end of the review period, I begged and borrowed the pair of D-250 Servos that JGH had reviewed in Vol.9 No.5. The first configuration was to use one D-250, driving the Calipers biwired as before from the 4 ohm taps. The dynamics were now increased over the Krell, but the lower midrange/upper bass veiling was more noticeable, being cloudy, and the upper midrange and lower treble became more forward in character, at the expense of the sense of depth. I know it's heresy, but there was a lack of coherence in the soundstage with the D-250 when compared with the Krell.

I then tried biamping the Calipers with both D-250s, but in the time I had available failed to get the results anticipated. The dynamics were, well, wonderful, the contrast between very soft and body-shaking loud approaching the live experience; the midrange became less forward than with one D-250, but there was still an overall lack of clarity. This was with one D-250 driving both bass panels and the other driving the tweeters. I scratched my head...Aha! I reached for my screwdriver. I drank it in one gulp and changed the wiring arrangements: now, each D-250 was driving a complete Caliper.

Better. Awesome, even. But still not right.

I will report more on this perplexing subject in a future issue.

Taking soundings
The last task before turning thoughts into printed words was to run a quick set of measurements. The impedance was more benign than that of the Scintilla: running between 8 and 14 ohms below 200Hz, it dropped to 3 ohms by 1kHz and rose gently to 6 ohms by 20k. Nothing there, unfortunately, that would explain the mismating with the D-250. The sensitivity appeared to be 4dB or so below that of the Celestion SL600, but this would be helped in-room by the typical dipole speaker's dispersion.

Taking a spatial average of the soundfield at the listening position with pink noise proved instructive: surprisingly for a panel speaker with such a small diaphragm—smaller, for example, than the Martin-Logan CLS—the 31.5Hz band was the highest in-room, the trend then being for the response gently—and very smoothly—to slope down all the way to 20kHz.

The fundamental woofer tuning of the earlier Apogees had quite a high Q, but Leo Spiegel has gone for a broader tuning with the Caliper. The woofer level is a little boosted, however, compared with the tweeter, which would explain why I ran into trouble with the deep organ notes on the Church Windows LP with the D-250. To ask for very high levels of ultra-low bass with an amplifier that has a rising output impedance at these frequencies will lead to a lack of control. The generous woofer balance will also allow bass problems to be clearly heard in the rest of the chain: I suspect that the upper-bass veiling was due, in part, to the LP12/Ittok's richness in that area.

Tonally, the Caliper's midrange and low treble were similar to the SL600, but above 3kHz there was less energy. Dull or recessed cartridges, or those with excessive low bass, would not be a good match with the Caliper. But use a suitable cartridge—the vdH MC10 would work well, I think—in a tonearm with superb bass definition, like the SME V, and these babies would really sing.

Measure for measure
"But this time with a little dedication..." sang Joan Armatrading in that "Love and Affection" track, and a little dedication is what you need to get the most from a speaker like the Caliper. Although it is amplifier-fussy—in the sense of requiring large amounts of voltage and amperage simultaneously— to a much lesser extent than the Scintilla, chameleon-like it manages to take on the sound character of the amplifier and cables with which it is used, to a larger extent than I have before experienced from a speaker.

Problem areas for the Caliper are the slight veiling in the upper bass/lower midrange (the extent of which will be very amplifier- and cable-dependent), the tendency for the low bass to be a little too rich, and the slight lack of HF air. Otherwise, the Calipers present the music with a midrange transparency rarely heard at this price level.

With a relatively neutral solid-state amplifier—I liked the Krell KSA-50, and Apogee also recommends the new $995 Belles 400, two Eagle 2As or Adcom GFA-555s—the Caliper is perfection on a small scale, exquisite if a little reticent above 120Hz, adequate below, transparent, but lacking bass authority and control.

There is a danger with a high-end loudspeaker like this, priced almost within the bounds of reality, in that owners of systems based on inexpensive front ends and electronics will be tempted to buy them. Certainly there should be no problems using the Calipers with relatively inexpensive solid-state amplification. However, they demand to be used in a system with an at-least-good LP player. The Linn is probably not a good choice; its own area of weakness, the upper bass, coincides with that of the Caliper. But SOTAs, VPIs, and Oracles would be fine.

With a compatible amplifier, the magic the Calipers weave on voice, their holographic, unforced, imagery, and the uncolored upper midrange and treble, add up to no problems with passing the goose-bump test.

Footnote 4: Apogee's specification indicates maximum levels of 105dB (C-weighted) will be attainable, which ties in with this level of bass. Apogee does say, however, that if the perforated plate supporting the magnets is very slightly bowed, bass excursion will be reduced. Current-production Calipers are checked for flatness.—John Atkinson



Sidebar: Specifications

Description: Full-range, two-way, ribbon loudspeaker. Frequency range: 30Hz to greater than 25kHz. Power requirements: 100W. Maximum spl: 105dB at 4m (C-weighting). Impedance: 3 ohms nominal.
Dimensions: 48" H x 24" W x 2" D. Weight: 70 lbs each.
Price: $1650/pair (1986); no longer available (2003).
Manufacturer: Apogee Acoustics Inc., 35 York Industrial Park, Randolph, MA 02368 (1986); Apogee Technology Inc., 129 Morgan Drive, Norwood, MA 02062. Tel: (781) 551-9450. Fax: (781) 440-9528—no longer manufacturing loudspeakers (2003). Web: Since Apogee no longer makes or services their speakers, readers can try for support information.