Magnepan MG2.6/R loudspeaker

91maggie26rAsk anyone in the street what they think of when they hear the word "loudspeaker" and odds are they'll describe a wooden box with moving-coil drivers sitting in its front. But ask a Stereophile reader and it's quite possible that he or she'll describe a large, flat panel reminiscent of a room divider: in our 1989 reader survey, the most widely represented brand of loudspeaker was Magnepan, with a significant lead over Infinity and Vandersteen, the second and third most common speaker brands. This represents considerable commercial success in a generally conservative marketplace for a company whose products are so different from the norm.

The subject of this review, the $1950/pair MG2.6/R, falls in the middle of Magnepan's range, above the Minnesota company's new quasi-ribbon models introduced at the 1991 June CES but below their top-line MG3.3 and Tympani designs. A two-way speaker, it uses a 40"-long version of Magnepan's ribbon tweeter, which features a ¼"-wide strip of corrugated aluminum foil, 2.5µm thick (1/10,000") hanging between two rows of magnets, with the edge clamped every 1.5" on opposite sides. The "Magneplanar" panel bass/midrange driver consists of wire conductors—in effect a flat voice-coil—bonded to a Mylar diaphragm positioned behind an array of bar magnets mounted on a steel supporting frame. This gives single-ended electromagnetic drive. Placing the diaphragm behind the magnets might be thought to obscure the sound, but the long vertical gaps between the strips of magnets should be acoustically transparent. The diaphragm is clamped at strategic points to minimize its resonant behavior.

Although it looks rather different, the 2.6 is basically similar to the MG2.5/R that J. Gordon Holt and I reviewed in Stereophile some three years ago. In fact, the ribbon unit is identical but the speaker's crossover has been extensively revised to give a higher sensitivity coupled with a warmer tonal balance. The crossover between the two drivers operates at approximately 1kHz; whereas the 2.5 used first-order electrical slopes, those in the 2.6 are second-order high-pass for the tweeter, which appears to be connected with inverted polarity, and third-order low-pass to the woofer/midrange. The crossover utilizes three air-core inductors (footnote 1), two paralleled plastic-film capacitors in series with the ribbon, and one non-polarized electrolytic capacitor shunting the woofer feed, these components somewhat untidily glued to the rear of the speaker's frame to the left of the input terminals. These are 4mm sockets with a locking hex-head screw. A second pair of 4mm sockets holds a shorting bar which can be replaced by a low-value power resistor to attenuate the ribbon output if desired. The somewhat fragile ribbon (fragile, that is, with respect to a typical moving-coil dome tweeter) is also protected against overload by a series 2.5A fuse. Replacing the tweeter unit is no problem, however.

Both drivers are mounted in/supported by a fiberboard baffle; the radiation pattern is fundamentally that of a dipole, with low-frequency extension governed by the baffle size. (When the wavelength of the sound is equal to the baffle dimensions, cancellation between the direct sound and the backwave will roll off the low bass in a gentle, nonresonant manner.) Hardwood strips (light oak, painted black, or unfinished) run down the edges of the baffle, these 2.5" wide and not as narrow as those of the MG2.5; the replaceable grille cloth is available in off-white, black, and gray. The bolt-on feet offer good support and extend both in front of and behind the speaker.

Though its dipole nature means that it will excite room resonant modes less efficiently than a typical box speaker (footnote 2), a panel speaker needs more careful positioning with respect to the rear wall, due to the reflection of the out-of-phase backwave interfering with the direct sound. Setting up the MG2.6/Rs to sound at their best therefore took some time, and, as might be expected from a large panel speaker featuring side-by-side drive-units, getting the optimum balance between the midrange one side of the crossover point and the treble the other side was also tricky. Ultimately, I settled for using the speakers with the ribbons on their outside edges—ribbons on the inside edges gave a sound that was too bright—toed-in slightly to the listening seat so that I was sitting about 15° off the ribbon axis. Each panel was about a third of the way into the room from both rear and side walls, which meant that I was just over 8' from the speakers, rather closer than most listeners would sit. I therefore ended up placing brass German Acoustics cones under the ends of the MG2.6s' rear feet to tilt the speakers a little toward my listening chair, which places my ears 36" from the ground.

I very much enjoyed my time with the MG2.6/Rs. They offered a big-boned, accessible sound that was pretty true to whatever kind of music I played on them. This overall excellence should be remembered while I describe what they do into the various areas of audiophile performance.

Midrange: Both J. Gordon Holt and I felt the original MG2.5/R to sound a little cool. The 2.6 is quite different, having a warm tonal balance that could occasionally become too much of a good thing. The left-hand register of the piano on my recording of Anna Maria Stanczyk playing Chopin on the Stereophile Test CD, for example, was a little too thickened in absolute terms with the speakers driven by the solid-state Levinson amplifier. Substituting the tubed Audio Research amplifier gave even more lower-midrange emphasis, to the point where I felt the sound was too much different from what it should have been. Live pianos don't sound that warm. (But the warmth did add a frisson of believability to Gary Karr's somewhat out-of-tune double bass on his Albinoni/Giazotto Adagio arrangement, Seven Seas K28C-170.)

Footnote 1: One of these inductors on the righthand loudspeaker had become detached in shipping, but was still connected and functioning correctly.

Footnote 2: This can be heard very easily by standing outside the listening-room door with the speakers playing full-range music. The bass region sounds very much less lumpy in the bass with a pair of dipole radiators in the room than with conventional loudspeakers.

Magnepan Inc.
1645 Ninth Street
White Bear Lake, MN 55110
(612) 426-1645

monetschemist's picture

Such a fine example of using listening as a means of critical evaluation - the difference in staging on various recordings perhaps refuting the idea of better staging just being due to the dipole arrangement, or the comment "live pianos don't sound that warm".

This is core Stereophile good stuff. Thank you so much!

My only tiny complaint would be that, while understanding the principle of reviewing equipment sold via local dealers, I feel it's really too bad not to have this kind of wonderful critical light shone on more stuff that is sold direct-to-consumer, since for many of us that's the only way we'll know in advance about the merits or otherwise of one of those alternatives.

dial's picture

I had the III/A (and even their tonearm) and sold them in the 90's for diy high efficiency.

Josh Hill's picture

This old review is a model of what a review should be! I particularly enjoyed the technical description, and the way JA correlated it with his subjective impressions.