Alternative Speaker Technologies (Aug. 1980)

Home | Audio Magazine | Stereo Review magazine | Good Sound | Troubleshooting

Departments | Features | ADs | Equipment | Music/Recordings | History

by Gary Stock

The time is the mid-'30s, and as Mussolini raves and war clouds gather over Europe, a British professor named MacLachlan is putting together one of the first texts on a new and interesting if minor device--the loudspeaker.

Sensibly enough, the book will be called Loud-speakers (how the English love their hyphens). MacLachlan does not have much prior art on which to build, since the direct-radiator loudspeaker debuted a scant nine years earlier. But in that brief period, he has managed to collect information on a reasonable array of sound-producing gizmos. He describes conventional cone-type loudspeakers with cones made of doped cloth, an amazing lightweight metal called duralumin (now known by the prosaic name "aluminum"), and a fairly new and exotic plastic called phenolic. He briefly covers the crystal earphone, describing how some minerals flex when an electric current passes through them, and can therefore be used to move small diaphragms and produce sound--but that is old hat to him. More space is devoted to a new idea called the con denser loudspeaker, a device com posed of two plates separated by a thin layer of felt. One moves, the other doesn't, he notes, and though that in itself suggests a certain amount of intrinsic distortion, MacLachlan is hopeful that the bugs can be worked out.

There is a short discussion, too, of a German wunder-lautensprecher called the Blatthaller, which consists of a supported rigid plate with a voice-coil zigzagging across the back through a carefully complementary zigzag arrangement of magnets. It is said to sound remarkably lifelike. With Scottish thoroughness, MacLachlan documents, diagrams, annotates, describes all of these devices, carefully but diplomatically details their strengths and weaknesses, and then makes discreet predictions as to their future value.

Forty-five years later, we are finally entering a new age in loudspeaker design--an age in which the supremacy of the technically untidy but ubiquitous moving-coil loudspeaker is for the first time being strongly challenged by the exotic transducer types described by MacLachlan. It has taken this long because only recently have the materials and processes--high energy magnets, plastic films, vapor-deposition processes, etc.--necessary to produce reasonably priced versions of these "alternative transducers" be come widely available. But every one of these concepts, as well as a few genuinely new wrinkles, is finally coming to the marketplace in some commercial form. Today's speaker engineer has more practical, interesting, and innovative options than at any time in the past 40 years.

This same range of options, of course, is also available to buyers of loudspeakers, and in order to help clarify the new choices that speaker purchasers have, we have assembled this guide to the transducer types. It is divided into six sections, including a review of the moving-coil principle for reference purposes. The distinctions between the various classes of transducer are not always clear ones, and some commercial examples, in fact, neatly straddle the line between one type and another. Our classifications are therefore only approximate, and our commentary on the merits and disadvantages of each is generally but not universally applicable. For those with an interest in the detail of the engineering techniques behind these concepts, we also provide a bibliography, broken down by transducer type.

--------- Conventional moving-coil speaker.

The Moving-Coil Driver--also called the dynamic, electrodynamic, or cone-type driver.

Like the internal combustion engine, the moving-coil loudspeaker is a Rube Goldberg-esque idea that has been made to work by dint of 50 years of painstaking twiddling and tweaking of its innards. Yet work it does, and sometimes very well. In terms of simple theory, this type of driver is a sort of fast-moving solenoid attached to a larger, flattened diaphragm that may be planar, dome or bowl shaped, ring like, or most commonly, a shallow conical frustrum. As detailed in the ac companying diagram, current flow through a tubular coil, which is immersed in a ring-shaped magnetic field, generates longitudinal motion; the attached diaphragm couples this motion to the air and induces the pressure variations that constitute sound. A number of problems are obvious. In the magnetic circuit, the field produced by the magnetic structure is uneven due to various fringing effects; either the coil or the magnetic gap must be longer than its mate in order to linearly accommodate the range of the driver's motion. On the diaphragm side of the device, one obvious difficulty lies in making the entire diaphragm surface precisely follow the coil's motion, particularly since the rather low efficiency of the magnetic motor system requires that the total moving mass of the speaker be kept low. Most of the "black art" of moving-coil speaker design, in fact, revolves around the details of cone shapes, paper mixtures, and adhesives to minimize the effects of nonlinear cone motion. The vast majority of the irregularities found in any loudspeaker response curve are due to dozens of minor resonances, nodes, anti-resonances, and phase cancellations taking place as the radiating surface wobbles back and forth.

Despite its difficulties, the conventional cone-type loudspeaker has been developed to a fairly high level. Motor systems that compensate for magnetic irregularities, synthetic diaphragm materials formed into flat plates or constructed in rigid sandwich-type composites, and system designs with en closures and dividing networks which compensate for many of the time-domain irregularities of moving-coil units have all been investigated in re cent years, with good results. And the advent of heat-sinking magnetic fluids and high-temperature voice-coil assemblies have improved power handling capabilities of cone-type units to the point where they are the unquestioned driver format for very high-power applications.

The Electrostatic Driver-- also called the capacitor or condenser-type loudspeaker.

----------Electrostatic transducer.

As the name implies, electrostatic drivers employ the attractive and repulsive forces that exist between electrostatically charged surfaces in order to generate sound. The conventional arrangement, first developed in the '30s (shortly after the publication of MacLachlan's book) places a single thin-film diaphragm with a very high surface resistivity between two acoustically transparent, electrically conductive grids. A fixed high-voltage charge on the center membrane causes it to be attracted to the varying voltage on one of the outer grids while it is simultaneously repelled by the reversed voltage of the other grid, the two grids normally receiving out-of-phase musical signals from a high-voltage amplifier or from a step-up transformer. This arrangement, in which both grids act to maintain a constant force on the diaphragm regardless of its displacement from the center position, is called push-pull for obvious reasons. It was the improvement that brought the single-sided electrostatic drivers of Mac Lachlan's day, with their high intrinsic distortion due to varying driving force as the diaphragm moved closer or further from a single grid, into the realm of high fidelity.

The virtues and difficulties of the electrostatic concept are clear even in this simplified description of operation. The diaphragm, being uniformly driven and of low mass with little in the way of self-resonant properties, can be expected to exhibit piston-like performance with very little coloration over a fairly wide frequency range (limited typically at the low end by the fundamental resonance of the diaphragm as a whole, and at the high end by the inductance of the step-up transformer or the frequency response of the driving high-voltage amplifier).

By the same token, its excursion will be limited to the spacing between the diaphragm and the grids, and its out put by the dielectric properties of air, since an extremely high polarizing voltage will arc across the air gap be tween the two. These factors have traditionally worked in concert to limit the dynamic range of electrostatic units and to mandate that they have fairly large, and therefore narrowly-dispersed radiating surfaces, although a number of innovative solutions have been developed in recent years. One is Dayton-Wright's use of an envelope of gas having much higher dielectric strength than air surrounding the drive elements. Another is the Beveridge approach, in which a large diaphragm radiates into a sort of sectoral horn, terminating in a tall, narrow slot, thus making the system a line-source radiator. An even more unusual approach is the one employed by BTM in their ES Translator series, in which a single central grid is placed between two diaphragms (see bibliography for an early paper on this design format), eliminating the need for high polarizing voltages. Other electrostatic drivers on the horizon may include self-polarized electret-type drivers (already used in some headphones), if the permanent polarizing electret charge can be in creased to a higher level by new techniques.

The Planar-Dynamic Driver--also called the planiform, magneto-dynamic, or distributed coil driver.

An exceptionally attractive (to the speaker designer) synthesis of the strong points of cone-type dynamic and electrostatic driver types, the planar-dynamic driver type is perhaps the fastest-growing alternative driver type today, currently used in at least 30 commercial loudspeaker systems.

As shown in the accompanying figure, a film diaphragm is suspended be tween two perforated or slotted mag net arrays, with a pattern of conductive lines on the diaphragm surface ar ranged so that the leakage flux from the magnet gaps drives the diaphragm in a push-pull arrangement (single-ended versions do exist, however). The driving surface is usually non-resonant, acoustically resistively loaded, like an electrostatic diaphragm, and more or less uniformly driven, depending on the density of the signal-carrying pat tern of lines. The load on the amplifier is stable, in contrast to both moving- coil and electrostatic formats, and presuming the use of a high-temperature plastic, and a fairly thick copper or aluminum pattern, the power handling can be made fairly high. Since the diaphragm mass is usually low, efficiency is controlled largely by the flux density of the moving system--and therefore in turn by the distance between the magnet structure and the diaphragm, as well as the energy product of the magnets themselves. This has led to the use of samarium cobalt "rare earth" magnets in many of the current examples of the genre, although units using conventional alnico and ferrite magnet materials have also been made. Interesting wrinkles in the first generation of designs include Strathern's use of non-driven film sheets adjacent to the diaphragm as a means of damping resonances, and Cerwin Vega's SUFT-FET drivers, which em ploy a single spiral track that covers al most the entire diaphragm surface. On the horizon are less expensive large-diaphragm versions, with area reduction as frequency rises to maintain good dispersion, and narrow strip-style full-range planar-dynamics with line-source dispersion characteristics.

The Ribbon Driver-- also called the leaf driver.

An approach used by only a handful of manufacturers in past years--Fane and Decca were for decades the only keepers of the ribbon faith--the rib bon has recently received a great deal of research and support in the Far East, as well as the support of a number of adherents in this country. One of the simplest transducer types, a ribbon consists of a narrow conductive diaphragm running lengthwise through a slot-shaped magnetic field, the diaphragm usually consisting of either an aluminum foil or a metalized plastic like aluminized mylar. The motion of the diaphragm is lengthwise, and the entire radiating surface consists of the ribbon's front and rear surfaces. Ad vantages include uniform drive and a consequent lack of diaphragm break up and resonance, extended high-frequency response, and (generally) good horizontal dispersion at high frequencies due to the narrowness of the radiating surface. Efficiency is generally very limited because of the fairly low flux densities permitted by the wide magnet gap and the small radiating area, and the excursion and power handling capabilities of the format are also usually fairly low. Among the de sign techniques commonly used to improve performance are horn loading of the diaphragm as a means of improving efficiency, pleating of the diaphragm as a means of increasing maxi mum excursion capability (the pleats "unfold," as it were, at peak amplitudes) and diaphragm rigidity, and the use of matching transformers to match the low impedance of the diaphragm--often only a fraction of an ohm--to the requirements of conventional amplifiers.

----Ribbon transducer.

----Heil "Air Motion Transformer" (courtesy ESS).

Among the interesting variations on the design type, Pyramid has developed an arrangement of rubber strands that fasten to the edges of the diaphragm and prevent twisting and non-pistonic motion, while Impulse has a wide-range ribbon system in which a metalized plastic strip several feet tall is used to generate a dipolar, line-source sound field. Another innovative variation, and one that has enjoyed enormous commercial success, is the Heil Air Motion Transformer, a system in which a folded plastic sheet with a ribbon attached to it operates by compressing and expanding along the folds in accordion-like fashion, thus generating high output levels (see the accompanying figure). The Heil arrangement thus avoids many of the difficulties of the traditional ribbon.

The Piezo-Electric Driver also called the piezo-ceramic or HPM driver.

----------Piezo-electric horn driver element (courtesy Motorola).

----------Cylindrical high-polymer piezo electric treble unit (courtesy Pioneer).

The piezo-electric principle is an ex ample of a phenomenon that appears mechanically simple while being extremely complex at the molecular level. In simple terms, piezo-electric substances are materials that move in some fashion when an electric current passes through them (the motion may be in one of two modes; physicists call the two types "twisters" and "benders"; the types of motion are self-explanatory). They include minerals like Rochelle salts, man-made ceramics like barium titanate (used in underwater sound projection, among other things), and some plastics belonging to the same general family as grocery-store meat wrapping, called the polyvinylidene fluorides (meat wrapping, in fact, is slightly piezo-electric, but not enough so for high-fidelity use).

Two firms have done most of the developmental work in piezo-electrics--Motorola, which manufactures a line of OEM treble drivers using thin wafers of ceramic material fastened to cones, and Pioneer, which manufactures a number of speaker drivers using sheets of polyvinylidene fluoride formed into cylindrical shapes, as well as flat headphone drivers, all under the trade name "HPM" (for High Polymer Material). Both types of units have characteristic falling electrical impedances and high mechanical impedances, and are therefore suitable for use at present only in high-frequency applications. The falling reactance, however, is fortuitously usable as a simple internal crossover element, and many of the Motorola drivers used in commercial loudspeaker systems are connected with only a simple power-limiting series resistor. Power handling in both types is fairly high, since there is little heat generated under normal conditions of use (a more common cause of failure under stress, notes a Motorola engineer in one paper, is fracturing of the thin, brittle ceramic elements). The Motorola units find frequent use in high-power disco and sound-reinforcement applications, in fact. On the horizon are wider range units, perhaps in the form of pulsating cylinders or dipolar strip-type arrays.


On the outer fringes of the loudspeaker design field are a number of other types, most notably the pricey ($6500 per pair) and complex Plasmatronics loudspeaker, which employs a midrange-treble driver consisting of an air-helium plasma (plasma in this sense indicating a gaseous substance with an abnormally high number of ions and a consequent charge) that thermally expands and contracts rap idly to produce sound. In the case of the Plasmatronics unit, the plasma is a lavender triangle-shaped mass fed by a high-power tube amplifier, with helium bled into the driver from a tank that must be refilled at regular intervals. The speaker is similar in many ways, according to inventor Alan Hill, to the well-known DuKane lonovac of the '60s, which at that time was beieved to operate on an entirely different principle (a thought-provoking comment on engineering as an art).

Other exotica include arc-discharge ("singing arcs") loudspeakers of the type demonstrated by one high-end amplifier manufacturer at a recent Consumer Electronics Show, although no commercial examples of this format now exist.

As MacLachlan's work aptly indicates, there is little that is truly new under the sun in any mechanical field--loudspeakers included, so most current engineers, including those actively involved in "blue-sky" development projects, don't see much in the way of truly revolutionary introductions over the next several years. Rather, the new age loudspeakers we've discussed will benefit in an evolutionary way from the availability of new materials and new measurement techniques, per haps eventually supplanting the moving-coil driver in high fidelity applications. The enormous potential of these new ideas, however, even in their barely developed present state, suggests we have many improvements to look forward to in the coming decade. Indeed a new age for speaker designers, and for listeners as well.


The drawings used in this article are based on those in Martin Colloms' excellent book High Performance Loudspeakers, published by Halsted Press, div. of John Wiley & Sons, 605 Third Ave., New York, N.Y. 10016.


Further Reading on the New Age Loudspeakers


Electroacoustics, F. V. Hunt, Wiley--The original and to date most complete examination of the electrostatic principle.

"Wide Range Electrostatic Loudspeaker," P. J. Walker, Wireless World, May-July, 1955--Design treatise on the first of the truly practical commercial electrostatic units, the Quad. Fascinating in its coverage of design de tails, especially because the Quad re mains one of the world's best today.

"Full-Range Electrostatic Loudspeakers," H. Leak and A. Sarker, Wireless World, October, 1956--First cover age of the "inside-out" electrostatic.

"A Wide Range Electrostatic Loudspeaker," Charles Malme, Jour. of the Audio Eng. Soc., January, 1959--Thorough coverage of an interesting electrostatic handmade as a graduate project at M. I. T. out of kitchen and garden parts.

"Electrostatic Loudspeaker Development," Arthur Janszen, Jour. of the Audio Eng. Soc., April, 1955--Practical examination of electrostatic design problems by one of the giants in the field.


High Performance Loudspeakers, Mar tin Colloms, Halsted Press--Comparative evaluation of the various drive mechanisms available to the de signer.

"The Isodynamic Principle," G. P. Millward, Proceedings of the Audio Eng. Soc., 50th Convention--Thorough going study of the planar-dynamic principle and its application to head phones, by a Wharfedale engineer.

"Electrodynamic Speaker Has Totally Active Surface," Stanley Rich, Electronics, June 16, 1961--A Bogen three way system of the early '60s with not only a planar-dynamic midrange and tweeter but also a pneumatically driven bass driver is analyzed by its developer.

"A Novel Planiform Loudspeaker Sys tem," R. Whelan, Proceedings of the Audio Eng. Soc., 50th Convention--Coverage of the innovative but ill fated Strathern wide-range planar-dynamic.


"Tweeter Using New Structure and New Material for Diaphragm (Direct-drive Ribbon Tweeter)," H. Nakajima, et al., Proceedings of the Audio Eng. Soc., 63rd Convention--Sony's rib bon tweeter, which straddles the line between planar-dynamic and ribbon, is discussed.

"Ribbon Velocity Microphones," Harry Olson, Jour. of the Audio Eng. Soc., June, 1970--The whole principle turned around and studied by one of the pioneers of loudspeaker design.

"Wide-range High-power Tweeter Using the Printed Planar Voice Coil (The Leaf Tweeter)," N. Sakamoto, et al., Proceedings of the Audio Eng. Soc., 58th Convention--Another Japanese ribbon unit covered in detail.


"Electrostatic Transducer with Piezo electric High-polymer Film," M. Tamura, et al., Jour. of the Audio Eng. Soc., January, 1975--Pioneer's design team discusses the chemistry and physics of their HPM drivers.

"A New Type of Tweeter Horn Employing a Piezoelectric Driver," Jonathan Bost, Jour. of the Audio Eng. Soc., October, 1975--Motorola's piezo-electric drivers are described in detail.


"The Ionic Loudspeaker for the Reproduction of High Frequencies," A. Falkus, British Kinematography Sound & Television Soc., Vol. 48, 1966-A theoretical treatment of the ionic approach.

"The Corona Wind Loudspeaker," G. Shirley, Jour. of the Audio Eng. Soc., Vol. 5, 1957--Somewhat similar in principle to the above, the title is one of the best teases in technical literature.

(adapted from Audio magazine, Aug. 1980)

Also see:

Some Loudspeakers Past and Present (Apr. 1970)

That Damping Factor (by Paul W. Klipsch) (Mar. 1970)

The Loudspeaker as a Spherical Sound Source (Mar. 1973)

Layman's Guide to LOUDSPEAKER SPECIFICATIONS--Part 3 (conclusion) (Jan. 1970)

= = = =

Prev. | Next

Top of Page    Home

Updated: Thursday, 2019-06-20 17:56 PST