Excerpted and adapted from The Complete Guide to High-End Audio, fifth edition. Copyright © 1994–2020 by Robert Harley. To order call (800) 841-4741. hifibooks.com
Many mechanisms for making air move in response to an electrical signal have been tried over the years. Three methods of creating sound work well enough—and are practical enough—to be used in commercially available products. These are the dynamic driver, the ribbon transducer, and the electrostatic panel. A loudspeaker using dynamic drivers is often called a box loudspeaker because the drivers are mounted in a box-like enclosure or cabinet. Ribbon and electrostatic loudspeakers are called planar loudspeakers because they’re usually mounted in flat, open panels.
The Dynamic Driver
The most popular loudspeaker technology is without question the dynamic driver. Loudspeakers using dynamic drivers are identifiable by their familiar cones and domes. The dynamic driver’s popularity is due to its many advantages: wide dynamic range, high power handling, high sensitivity, relatively simple design, and ruggedness. Dynamic drivers are also called point-source transducers because the sound is produced from a point in space.
Dynamic loudspeaker systems use a combination of different-sized dynamic drivers. The low frequencies are reproduced by a cone woofer. High frequencies are generated by a tweeter, usually employing a small metal or fabric dome. Some loudspeakers use a third dynamic driver, the midrange, to reproduce frequencies in the middle of the audio band.
Despite the very different designs of these drivers, they all operate on the same principle (Fig. 1). First, the simplified explanation: Electrical current from the power amplifier flows through the driver’s voice coil. This current flow sets up a magnetic field around the voice coil that expands and collapses at the same frequency as the audio signal. The voice coil is suspended in a permanent magnetic field generated by magnets in the driver. This permanent magnetic field interacts with the magnetic field generated by current flow through the voice coil, alternately pushing and pulling the voice coil back and forth. Because the voice coil is attached to the driver’s cone, this magnetic interaction pulls the cone back and forth, producing sound.
On a more technical level, the voice coil is a length of wire wound around a thin cylinder called the voice-coil former. The former is attached to the diaphragm (a cone or dome). Electrical current from the amplifier flows through the voice coil, which is mounted in a permanent magnetic field whose magnetic lines of flux cross the gap between two permanent magnets. According to the “right-hand rule” of physics, the circular flow of current through the voice-coil windings generates magnetic forces that are directed along the voice coil’s axis. The interaction between the fluctuating field of the voice coil and the fixed magnetic field in the gap produces axial forces that move the voice coil back and forth, carrying the diaphragm with it. The faster the audio signal alternates, the faster the diaphragm moves, and the higher the frequency of sound produced. Dynamic drivers are also called moving-coil drivers, for obvious reasons.
Other elements of the dynamic driver include a spider that suspends the voice coil in place as it moves back and forth. The basket is a cast- or stamped-metal structure holding the entire assembly together. (Cast baskets are generally found in higher-quality loudspeakers, stamped baskets in budget models.) A ring of compliant rubber material, called the surround, attaches the cone to the basket rim. The surround allows the cone to move back and forth while still attached to the basket. The maximum distance the cone moves back and forth is called its excursion.
Common cone materials include paper, paper impregnated with a stiffening agent, a form of plastic such as polypropylene, metal (titanium, for example), or exotic materials such as carbon-fiber, Kevlar (the material used in bullet-proof vests), and proprietary composites. Designers use these materials, and sometimes a sandwich of different materials, to prevent a form of distortion called breakup. Breakup occurs when the cone material flexes instead of moving as a perfect piston. Because the cone is driven at a small area at the inside (an area the size of the voice coil), the cone tends to flex, which produces non-linear distortion. Stiff cone materials help prevent breakup. Although all dynamic drivers exhibit breakup at a certain frequency, the competent loudspeaker designer makes sure that a driver is never driven by frequencies that would cause breakup. For example, if a woofer has its first breakup mode at 4kHz, the designer would probably operate the driver only up to about 2kHz, well away from the breakup frequency.
The cone should be lightweight as well as stiff. A lighter cone has less inertia, allowing it to respond faster to transient signals and stop faster after the drive signal has ceased. Think of a bass-drum whack. A large and heavy cone may not be able to move fast enough to reproduce the sound’s attack, diminishing the drum’s dynamic impact. Similarly, once the drum whack is over, the heavy cone’s mass will want to continue moving. Loudspeaker designers therefore search for cone materials that combine high stiffness with low mass. Many of the advances made in loudspeaker design of the past 15 years have been the result of materials research that has yielded lighter yet stiffer diaphragms.
Tweeters work on the same principle, but typically use a 1" dome instead of a cone. Common dome materials include plastic, woven fiber coated with a rubbery material, titanium, aluminum and aluminum alloys, and gold-plated aluminum. A recent trend has been toward making tweeter diaphragms from materials such as beryllium and even diamond. These materials, once found only in a company’s expensive flagship models, have found their way into less costly products. Unlike cone drivers, which are driven at the cone’s apex, dome diaphragms are driven at the dome’s outer perimeter. Some dome tweeters use Ferrofluid, a liquid cooling agent, to remove heat from the tweeter’s small voice coil. The first breakup mode of well-designed modern dome tweeters is above 25kHz, well out of the audible range.
Midrange drivers are smaller versions of the cone woofer. Some, however, use dome diaphragms instead of cones.
When a dynamic driver plays loudly, it must dissipate a tremendous amount of heat from its voice coil. As the voice coil heats up, its electrical resistance increases, reducing the amount of current flow through the voice coil. Because it is current flow through the voice coil that causes the cone to move back and forth, increased voice-coil resistance reduces the amount of sound produced by the driver. In other words, you can keep turning up the volume, but the speaker reaches a point where it won’t play any louder.
This phenomenon, called dynamic compression, obviously alters musical dynamics. Specifically, loud musical peaks aren’t reproduced quite as loudly as they should be. An electroacoustic mechanism is changing the score’s dynamic markings; fff may be reproduced as ff, for example.
Although this distortion of the music’s dynamics is a departure from accuracy, dynamic compression introduces a worse form of distortion. The multiple drivers in a three-way dynamic loudspeaker will have different levels at which they begin to compress. If the tweeter exhibits dynamic compression at a lower volume than the midrange and woofer, the sound will be slightly duller on musical peaks. Conversely, if the woofer and midrange compress at a lower level than the tweeter, the sound will grow brighter as it gets louder. In essence, the speaker’s tonal balance changes as a function of the music’s dynamics. You may not hear this phenomenon directly as a change in tonal balance, but this form of distortion creates an impression of strain on musical peaks that may momentarily draw our attention away from the music. It’s just one more cue to the brain that we’re not hearing live music.
Loudspeaker designers combat dynamic compression by creating drivers with oversized voice coils, physical structures that naturally cool the voice coil, and through innovative voice-coil wire shape and winding techniques.
There’s a related mechanism by which drivers can distort music. When a driver is producing a loud sound, the ends of its voice coil can momentarily leave the magnetic field in which the coil sits. This robs the driver of its ability to faithfully move back and forth in response to the drive signal, again changing the music’s dynamics and creating distortion. This problem is addressed by designing drivers with a very short voice coil and a very long magnetic gap. Although a short voice coil is more susceptible to dynamic compression, it doesn’t introduce this form of non-linear distortion. Loudspeaker designers must balance these trade-offs.
Finally, a dynamic driver can “bottom out,” heard as a popping or cracking sound from the woofer. The popping sound is caused by the voice-coil former (the bobbin around which the voice coil is wound) hitting the back of the magnet structure. If you ever hear this sound, immediately turn down the volume.
Problems with Dynamic Drivers
Of all the components in your system, the loudspeaker is the most likely to develop trouble. The most common problem is a blown driver, usually the tweeter—it simply stops working. Tweeters are often destroyed by too much current flowing through their voice coils. The tweeter can’t dissipate the resultant heat quickly enough, and its voice coil is burned open.
Another common cause of speaker failure is the buzzing sound produced by a loose voice coil. Too much current through the voice coil melts the glue holding the coil to the former. This loosens turns of wire, which then rub against the magnet to cause the buzzing sound.
Driver mounting bolts can loosen over time and degrade a loudspeaker’s performance by allowing the entire driver to vibrate rather than just its diaphragm. Gently tightening these bolts from time to time—particularly when a loudspeaker is new—can improve its sound. Be careful not to overtighten and strip the bolts.
The Electromagnetic Dynamic Driver
The dynamic driver I’ve described eventually runs up against the laws of physics. Specifically, the magnetic field strength generated by the fixed magnets is limited, which in turn places restrictions on the cone weight, how low in frequency the driver will play, and how sensitive the driver is. A heavy cone goes lower in frequency (all other factors being equal), but requires greater magnetic-field strength surrounding the voice coil to drive it. These limitations are particularly acute in large woofers.
A solution to this physics problem is the electromagnetic driver (also called a field-coil driver) in which the driver’s fixed magnets are replaced with a large coil that functions as an electromagnet. The coil is driven with direct current from an outboard power supply that plugs into an AC outlet. This current flow through the coil creates the magnetic field against which the voice-coil–generated magnetic field pushes and pulls. The electro-magnet produces a magnetic field strength in the gap (the area in which the voice coil sits) nearly double that of a conventionally driven woofer. Consequently, the electromagnetic woofer can be heavier (giving it a lower resonant frequency) yet simultaneously more efficient. Moreover, the woofer’s bass output can be adjusted by varying the current through the electromagnetic coil. This is accomplished via a control on the outboard supply that drives current through the electromagnetic coil. One can thus adjust the electromagnetic woofer’s bass output to better integrate the system into a variety of listening rooms.
An example of the electromagnetic driver is the woofer in the Focal Grande Utopia EM loudspeaker. The system’s woofer has a very high sensitivity (97dB for 1W) but very low resonance (24Hz). In other words, the woofer delivers lots of very low bass with very little input power. The price of this performance is the need for the outboard supply that has to be plugged into an AC outlet, along with the sheer weight of the woofer. The EM’s 16" woofer weighs 63 pounds, 48 of which is the electromagnetic coil.
The Planar-Magnetic Transducer
The next popular driver technology is the planar-magnetic transducer, also known as a ribbon driver. Although the term “ribbon” and “planar-magnetic” are often used interchangeably, a true ribbon driver is actually a sub-class of the planar-magnetic driver. Let’s look at a true ribbon first.
Instead of using a cone attached to a voice coil suspended in a magnetic field, a ribbon driver uses a strip of material (usually aluminum) as a diaphragm suspended between the north-south poles of two magnets (see Fig. 2). The ribbon is often pleated for additional strength. The audio signal travels through the electrically conductive ribbon, creating a magnetic field around the ribbon that interacts with the permanent magnetic field. This causes the ribbon to move back and forth, creating sound. In effect, the ribbon functions as both the voice coil and the diaphragm. The ribbon can be thought of as the voice coil stretched out over the ribbon’s length.
In all other planar-magnetic transducers, a flat or slightly curved diaphragm is driven by an electromagnetic conductor. This conductor, which is bonded to the back of the diaphragm, is analogous to a dynamic driver’s voice coil, here stretched out in straight-line segments. In most designs, the diaphragm is a sheet of plastic with the electrical conductors bonded to its surface. The flat metal conductor provides the driving force, but it occupies only a portion of the diaphragm area. Such drivers are also called quasi-ribbon transducers. Fig. 3 shows the difference between a true ribbon and a quasi-ribbon driver.
A planar driver is a true ribbon only if the diaphragm is conductive and the audio signal flows directly through the diaphragm, rather than through conductors bonded to a diaphragm, as in quasi-ribbon drivers. (Despite this semantic distinction, I’ll use the term “ribbon” throughout the rest of this section, with the understanding that it covers both true ribbons and quasi-ribbon drivers.)
Ribbon drivers like the one in Figs. 2 and 3 are called line-source transducers because they produce sound over a line rather than from a point, as does a dynamic loudspeaker. Moreover, a ribbon’s radiation pattern changes dramatically with frequency. At low frequencies, when the ribbon’s length is short compared to the wavelength of sound, the ribbon will act as a point source and produce sound in a sphere around the ribbon—just like a point-source woofer. As the frequency increases and the wavelength of sound approaches the ribbon’s dimensions, the radiation pattern narrows until it looks more like a cylinder around the ribbon than a sphere. At very high frequencies, the ribbon radiates horizontally but not vertically. This can be an advantage in the listening room: the listener hears more direct sound from the loudspeaker and less reflection from the sidewalls and ceiling. Reduced wall reflections aid in soundstaging: Pinpoint imaging and the ability to project a concert hall’s acoustic signature are hallmarks of good ribbon loudspeakers.
The main technical advantage of a ribbon over a moving-coil driver is the ribbon’s vastly lower mass. Instead of using a heavy cone, voice coil, and voice-coil former to move air, the only thing moving in a ribbon is a very thin strip of aluminum. A ribbon tweeter may have one tenth the mass and 10 times the radiating area of a dome tweeter’s diaphragm. Low mass is a high design goal: the diaphragm can respond more quickly to transient signals. In addition, a low-mass diaphragm will stop moving immediately after the input signal has ceased. The ribbon starts and stops faster than a dynamic driver, allowing it to more faithfully reproduce transient musical information.
The ribbon driver is usually mounted in a flat, open-air panel that radiates sound to the rear as well as to the front. A loudspeaker that radiates sound to the front and rear is called a dipole, which means “two poles.” Fig. 4 shows the radiation patterns of a point-source loudspeaker (left) and a dipolar loudspeaker.
Another great advantage enjoyed by ribbons is the lack of a box or cabinet. As we’ll see in the section of this chapter on loudspeaker enclosures, the enclosure can greatly degrade a loudspeaker’s performance. Not having to compensate for an enclosure makes it easier for a ribbon loudspeaker to achieve stunning clarity and lifelike musical timbres.
A full-range quasi-ribbon loudspeaker is illustrated in Fig. 5. The large panel extends the system’s bass response: when the average panel dimension approaches half the wavelength, front-to-rear cancellation reduces bass output. Consequently, the larger the panel, the deeper the low-frequency extension.
Ribbon loudspeakers are characterized by a remarkable ability to produce extremely clean and quick transients—such as those of plucked acoustic guitar strings or percussion instruments. The sound seems to start and stop suddenly, just as one hears from live instruments. Ribbons sound vivid and immediate without being etched or excessively bright. In addition, the sound has an openness, clarity, and transparency often unmatched by dynamic drivers. (Incidentally, these qualities are shared by ribbon microphones.) Finally, the ribbon’s dipolar nature produces a huge sense of space, air, and soundstage depth (provided this spatial information was captured in the recording). Some argue, however, that this sense of depth is artificially produced by ribbon loudspeakers, rather than being a reproduction of the actual recording.
Despite their often stunning sound quality, ribbon drivers have several disadvantages. The first is low sensitivity; it takes lots of amplifier power to drive them. Second, ribbons inherently have a very low impedance, often a fraction of an ohm. Most ribbon drivers therefore have an impedance matching transformer in the crossover to present a higher impedance to the power amplifier. Design of the transformer is therefore crucial to prevent it from degrading sound quality.
From a practical standpoint, ribbon-based loudspeakers are more difficult to position in a room. Small variations in placement can greatly change the sound, due primarily to their dipolar radiating pattern. This dipolar pattern requires that the ribbon loudspeakers be placed well away from the rear wall, and that the rear wall be acoustically benign.
Low-profile, ribbon-based loudspeakers with the ribbon top at the same height as the listener’s ears will have a radically different treble balance if the listener moves up or down a few inches. That’s because ribbon loudspeakers have very narrow vertical dispersion, meaning they radiate very little sound above and below the ribbon at high frequencies. If you sit too high or listen while standing, you’ll hear less treble. Some ribbon loudspeakers have a tilt adjustment that allows you to set the correct treble balance for a particular listening height.
Ribbons also have a resonant frequency that, if excited, produces the horrible sound of crinkling aluminum foil. Consequently, the ribbon must be used within strict frequency-band limits. In addition, ribbon drivers are tensioned at the factory for optimum performance. If under too much tension, a ribbon will produce less sound; if under too little tension, a ribbon can produce distortion that sounds like the music is “breaking up.” This is most noticeable on piano; the transient leading edges sound “shattered” and distorted. A sudden increase in ambient temperature can cause a ribbon driver to lose some of its tension and introduce the distortion described. If you hear this sound from your ribbon loudspeakers, contact the manufacturer for advice. The solution may be as simple as turning a few tensioning bolts half a turn.
Ribbon drivers don’t necessarily have to be long and thin. Variations on ribbon technology have produced drivers having many of the desirable characteristics of ribbons but few of the disadvantages.
Finally, some loudspeakers use a combination of dynamic and ribbon transducers to take advantage of both technologies. These so-called hybrid loudspeakers typically use a dynamic woofer in an enclosure to reproduce bass, and a ribbon midrange/tweeter. The hybrid technique brings the advantages of ribbon drivers to a lower price level (ribbon woofers are big and expensive), and exploits the advantages of each technology while avoiding the drawbacks. The challenge in such a hybrid system is to achieve a seamless transition between the dynamic woofer and the ribbon tweeter, with no audible discontinuity between the drivers.
The Heil Air Motion Transformer
Designed in the early 1970s by Dr. Oskar Heil, the Heil Air Motion transformer operates in a completely different way than dynamic or planar drivers. Rather than causing a diaphragm to move back and forth like a piston, the AMT’s pleated polyethylene membrane alternately squeezes and expands the air in front of it in response to the audio signal. A conductor is bonded to the diaphragm, and then suspended in a magnetic field. Although ribbons are often pleated for strength, the AMT’s pleats are much deeper and are responsible for the squeezing effect. The “Air Motion Transformer” name comes from the fact that air is squeezed out of the pleats at five times the speed of the diaphragm’s motion. This squeezing effect, and the accompanying acceleration of the air, is analogous to the rapid expulsion of a seed from a piece of fruit that’s gently squeezed. The seed’s motion is many times faster than the force applied.
The AMT has excellent transient response due to the diaphragm’s low mass and very small range of motion, as well as high acoustic output. A 1" wide AMT produces the same acoustic output as an 8" round driver.
A number of companies make a version of the Heil Air Motion Transformer now that the patents have expired.
The Electrostatic Driver
Like the ribbon transducer, an electrostatic driver uses a thin membrane to make air move. But that’s where the similarities end. While both dynamic and ribbon loudspeakers are electromagnetic transducers—they operate by electrically induced magnetic interaction—the electrostatic loudspeaker operates on the completely different principle of electrostatic interaction.
No discussion of electrostatic loudspeakers would be complete without mentioning the classic electrostatic loudspeaker, the QUAD ESL-57, created in 1957 by Peter Walker. The ESL-57 revolutionized the standard for transparency upon its introduction, and still holds it own more than 60 years later. Many listeners’ first exposure to high-quality audio was through an ESL-57. A large number of contemporary loudspeaker designers still have a pair of ESL-57s on hand as a reference. The ESL-57 doesn’t have much low bass, won’t play very loudly, and produces a very narrow sweet spot, but when operated within its limitations, it’s magical.
In an electrostatic loudspeaker (sometimes called an ESL), a thin moveable membrane—usually made of transparent Mylar—is stretched between two static elements called stators (Fig. 6). The membrane is charged to a very high voltage with respect to the stators. The audio signal is applied to the stators, which create electrostatic fields around them that vary in response to the audio signal. The varying electrostatic fields generated around the stators interact with the membrane’s fixed electrostatic field, pushing and pulling the membrane to produce sound. One stator pulls the membrane, the other pushes it. This illustration also shows a dynamic woofer as part of a hybrid dynamic/electrostatic system.
The voltages involved in an electrostatic loudspeaker are very high. The polarizing voltage applied to the diaphragm may be as high as 10,000 volts (10kV). In addition, the audio signal is stepped up from several tens of volts to several thousand volts by a step-up transformer inside the electrostatic loudspeaker. These high voltages are necessary to produce the electrostatic fields around the diaphragm and stators.
To prevent arcing—the electrical charge jumping between elements—the stators are coated with an insulating material. Still, if an electrostatic loudspeaker is overdriven, the electrostatic field strips free electrons from the oxygen in the air, making it ionized; this provides a conductive path for the electrical charge. Large diaphragm excursions—i.e., a loud playing level—put the diaphragm closer to the stators and also encourage arcing. Arcing can destroy electrostatic panels by punching small holes in the membrane. Arcing is a greater problem in humid climates than in dry climates because moisture makes the air between the stators more electrically conductive.
Electrostatic panels are often divided into several smaller panels to reduce the effects of diaphragm resonances. Some panels are curved to reduce the lobing effect (uneven radiation pattern) at high frequencies. Lobing occurs when the wavelength of sound is small compared to the diaphragm. Lobing is responsible for electrostatics’ uneven high-frequency dispersion pattern, which Stereophile magazine founder J. Gordon Holt has dubbed the vertical venetian-blind effect, in which the tonal balance changes rapidly and repeatedly as you move your head from side to side.
Electrostatic panels are of even lighter weight than planar- magnetic transducers. Unlike the ribbon driver, in which the diaphragm carries the audio signal current, the electrostatic diaphragm need not carry the audio signal. The diaphragm can therefore be very thin, often less than 0.001". Such a low mass allows the diaphragm to start and stop very quickly, improving transient response. And because the electrostatic panel is driven uniformly over its entire area, the panel is less prone to breakup. Both the electromagnetic planar loudspeaker (a ribbon) and the electrostatic planar loudspeaker enjoy the benefits of limited dispersion, which means less reflected sound arriving at the listening position. Like ribbon loudspeakers, electrostatic loudspeakers also have no enclosure to degrade the sound. Electrostatic loudspeakers also inherently have a dipolar radiation pattern. Because the diaphragm is mounted in an open panel, the electrostatic driver produces as much sound to the rear as to the front. Finally, the electrostatic loudspeaker’s huge surface area confers an advantage in reproducing the correct size of instrumental images.
In the debit column, electrostatic loudspeakers must be plugged into an AC outlet to generate the polarizing voltage. Because the electrostatic is naturally a dipolar radiator, room placement is more crucial to achieving good sound. The electrostatic loudspeaker needs to be placed well out into the room and away from the rear wall to realize a fully developed soundstage. Electrostatics also tend to be insensitive, requiring large power amplifiers. The load impedance they present to the amplifier is also more reactive than that of dynamic loudspeakers, further taxing the power amp. (Reactance is described later in this chapter.) Nor will they play as loudly as dynamic loudspeakers; electrostatics aren’t noted for their dynamic impact, power, or deep bass. Instead, electrostatics excel in transparency, delicacy, transient response, resolution of detail, stunning imaging, and overall musical coherence.
Electrostatic loudspeakers can be augmented with separate dynamic woofers or a subwoofer to extend the low-frequency response and provide some dynamic impact. Other electrostatics achieve the same result in a more convenient package: dynamic woofers in enclosures mated to the electrostatic panels. Some of these designs achieve the best qualities of both the dynamic driver and electrostatic panel. Audition such hybrid speakers carefully; they sometimes exhibit an audible discontinuity at the transition frequency at which the woofer leaves off and the electrostatic panel takes over. Listen, for example, for a change in a piano’s timbre, bloom, projection, and image size as it is played in different registers. Acoustic bass in jazz is also a good test of woofer/panel discontinuity in dynamic/electrostatic hybrid loudspeakers.
One great benefit of full-range ribbons and full-range electrostatics is the absence of a crossover; the diaphragm is driven by the entire audio signal. This prevents any discontinuities in the sound as different frequencies are reproduced by different drivers. In addition, removing the resistors, capacitors, and inductors found in crossovers greatly increases the full-range planar’s transparency and harmonic accuracy. Even hybrid planars put the crossover frequency between the dynamic woofer and the planar panel very low (below 800Hz, a frequency nearly an octave above middle A), so there’s no discontinuity between drivers through most of the audible spectrum.
Finally, the large diaphragms of electrostatic and ribbon drivers are gently driven over their entire surface areas, rather than forcefully over the relatively small voice-coil area of a dynamic driver. This high force over a small area contributes to the dynamic driver’s breakup described earlier, a phenomenon less likely to occur with large planar diaphragms.
Excerpted and adapted from The Complete Guide to High-End Audio, fifth edition. Copyright © 1994–2020 by Robert Harley. To order call (800) 841-4741. hifibooks.com