An electrostatic loudspeaker (ESL) is a loudspeaker design in which sound is generated by the force exerted on a membrane suspended in an electrostatic field.
Contents |
The speakers use a thin flat diaphragm usually consisting of a plastic sheet coated with a conductive material such as graphite sandwiched between two electrically conductive grids, with a small air gap between the diaphragm and grids. For low distortion operation, the diaphragm must operate with a constant charge on its surface, rather than with a constant voltage(charge and voltage are not the same thing). This is accomplished by either or both of two techniques: the diaphragm's conductive coating is chosen and applied in a manner to give it a very high surface resistivity, and/or a large value resistor is placed in series between the EHT (Extra High Tension or Voltage) power supply and the diaphragm (resistor not shown in the diagram here).However the latter technique will still allow distortion as the charge will migrate across the diaphragm to the point closest to the "grid" or electrode thereby increasing the force moving the diaphragm, this will occur at audio frequency so the diaphragm requires a high resistance (megohms) to slow the movement of charge for a practical speaker.
The diaphragm is usually made from a polyester film (thickness 2–20 µm) with exceptional mechanical properties, such as PET film. By means of the conductive coating and an external high voltage supply the diaphragm is held at a DC potential of several kilovolts with respect to the grids. The grids are driven by the audio signal; front and rear grid are driven in antiphase. As a result a uniform electrostatic field proportional to the audio signal is produced between both grids. This causes a force to be exerted on the charged diaphragm, and its resulting movement drives the air on either side of it.
In virtually all electrostatic loudspeakers the diaphragm is driven by two grids, one on either side, because the force exerted on the diaphragm by a single grid will be unacceptably non-linear, thus causing harmonic distortion. Using grids on both sides cancels out voltage dependent part of non-linearity but leaves charge (attractive force) dependent part.[1] The result is near complete absence of harmonic distortion. In one recent design, the diaphragm is driven with the audio signal, with the static charge located on the grids (Transparent Sound Solutions).
The grids must be able to generate as uniform an electric field as possible, while still allowing for sound to pass through. Suitable grid constructions are therefore perforated metal sheets, a frame with tensioned wire, wire rods, etc.
To generate a sufficient field strength, the audio signal on the grids must be of high voltage. The electrostatic construction is in effect a capacitor, and current is only needed to charge the capacitance created by the diaphragm and the stator plates (previous paragraphs referred to as grids or electrodes). This type of speaker is therefore a high-impedance device. In contrast, a modern electrodynamic cone loudspeaker is a low impedance device, with higher current requirements. As a result, impedance matching is necessary in order to use a normal amplifier. Most often a transformer is used to this end. Construction of this transformer is critical as it must provide a constant (often high) transformation ratio over the entire audible frequency range (i.e. large bandwidth) and so avoid distortion. The transformer is almost always specific to a particular electrostatic speaker. To date, Acoustat built the only commercial "transformer-less" electrostatic loudspeaker. In this design, the audio signal is applied directly to the stators from a built-in high-voltage valve amplifier (as valves are also high impedance devices), without use of a step-up transformer.
Advantages of electrostatic loudspeakers include levels of distortion one to two orders of magnitude lower than conventional cone drivers in a box, the extremely light weight of the diaphragm which is driven across its whole surface, and exemplary frequency response (both in amplitude and phase) because the principle of generating force and pressure is almost free from resonances unlike the more common electrodynamic driver. Musical transparency can be better than in electrodynamic speakers because the radiating surface has much less mass than most other drivers and is therefore far less capable of storing energy to be released later. For example, typical dynamic speaker drivers can have moving masses of tens or hundreds of grams whereas an electrostatic membrane only weighs a few milligrams, several times less than the very lightest of electrodynamic tweeters. The concomitant air load, often insignificant in dynamic speakers, is usually tens of grams because of the large coupling surface, this contributing to damping of resonance buildup by the air itself to a significant, though not complete, degree. Electrostatics can also be executed as full-range designs, lacking the usual crossover filters and enclosures that could color or distort the sound.
Since many electrostatic speakers are tall and thin designs without an enclosure, they act as a vertical dipole line source. This makes for rather different acoustic behavior in rooms compared to conventional electrodynamic loudspeakers. Generally speaking, a large-panel dipole radiator is more demanding of a proper physical placement within a room when compared to a conventional box speaker, but, once there, it is less likely to excite bad-sounding room resonances, and its direct-to-reflected sound ratio is higher by some 4–5 decibels. This in turn leads to more accurate stereo reproduction of recordings that contain proper stereo information and venue ambience. Planar (flat) drivers tend to be very directional giving them good imaging qualities, on the condition that they have been carefully placed relative to the listener and the sound-reflecting surfaces in the room. Curved panels have been built, making the placement requirements a bit less stringent, but sacrificing imaging precision somewhat.
Disadvantages include a lack of bass response (due to phase cancellation from a lack of enclosure (bass rolloff 3db point occurs when the narrowest panel dimension equals a quarter wavelength of the radiated frequency for dipole radiators, so for a Quad ESL 63 at 0.66 meters wide this occurs at around 129 Hz so is comparable to many box speakers. speed of sound taken as 343 m/s) and the difficult physical challenge of reproducing low frequencies with a vibrating taut film with little excursion amplitude, however as most diaphragms have a very large surface area compared to cone drivers only small amplitude excursions are required to put relatively large amounts of energy out), and sensitivity to ambient humidity levels. While bass is lacking quantitatively (due to lower distortion than cone drivers) it can be of better quality ('tighter' and without 'booming') than that of electrodynamic (cone) systems. Phase cancellation can be somewhat compensated for by electronic equalization (a so-called shelving circuit that boosts the region inside the audio band where the generated sound pressure drops because of phase cancellation). Nevertheless maximum bass levels cannot be augmented because they are ultimately limited by the membrane's maximum permissible excursion before it comes too close to the high-voltage stators, which may produce electrical arcing and burn holes through it. Recent, technically more advanced solutions for perceived lack of bass include the use of large, curved panels (Sound Lab, MartinLogan CLS), electrostatic subwoofer panels (Audiostatic, Quad) and long-throw electrostatic element allowing large diaphragm excursions (Audiostatic). Another trick often practised is to step up the bass (20–80 Hz) with a higher transformation ratio than the mid and treble.
This relative lack of loud bass is often remedied with a hybrid design using a dynamic loudspeaker, e.g. a subwoofer, to handle lower frequencies with the electrostatic diaphragm handling middle and high frequencies. Many feel that the best low frequency unit for hybrids are cone drivers mounted on open baffles as dipoles transmission line woofers or horns, since they possess roughly the same qualities (at least in the bass) as electrostatic speakers, i.e. good transient response, little box coloration, and (ideally) flat frequency response. However, there is often a problem with integrating such a woofer with the electrostatics. This is because most electrostatics are line sources, the sound pressure level of which decreases by 3 dB for each doubling of distance. A cone speaker's sound pressure level, on the other hand, decreases by 6 dB for each doubling of distance because it behaves as a point source. This can be overcome by the theoretically more elegant solution of using conventional cone woofer(s) in an open baffle, or a push-pull arrangement, which produces a bipolar radiation pattern similar to that of the electrostatic membrane. This is still subject to phase cancellation, but cone woofers can be driven to far higher levels due to their longer excursion, thus making equalization to a flat response easier and they add distortion thereby increasing the area (and therefore the power) under the frequency response graph, making the total low frequency energy higher but the fidelity to the signal lower.
The directionality of electrostatics can also be a disadvantage in that it means the 'sweet spot' where proper stereo imaging can be heard is relatively small, limiting the number of people who can fully enjoy the advantages of the speakers simultaneously.
Because of their tendency to attract dust, insects, conductive particles and moisture, electrostatic speaker diaphragms will gradually deteriorate and need periodic replacement. They also need protection measures to physically isolate their high voltage parts from accidental contact with humans and pets. Cost-effective repair and restoration service is available for virtually every current and discontinued electrostatic loudspeaker model.
Electrostatic speakers enjoy some popularity among do-it-yourself (DIY) loudspeaker builders. They are one of the few types of speakers in which the transducers themselves can be built from scratch by an amateur. Basic hardware for complete ESL DIY projects is available all over the web. Such supplies include resistors and capacitors for RC-circuit frequency equalization, if necessary; step-up transformers; perforated metal sheets or grids and insulating plastics for the stators; polymer film and conductive paint (e.g. a liquid graphite suspension) for the membrane; simple tensioning equipment for proper membrane tuning; and a frame, usually of wood, to hold everything together. A widely-read resource by ESL enthusiasts is The Electrostatic Loudspeaker Design Cookbook (ISBN 978-1-882580-00-2) by notable ESL specialist Roger Sanders.[2]
Arthur Janszen was granted U.S. Patent 2,631,196 in 1953 for an electrostatic loudspeaker. The developers of the Tri-Ergon sound-on-film sound film system had developed a primitive design of electrostatic loudspeaker as early as 1919. Mr. Janszen's company, JansZen still makes an evolved version of his original design. [3]
The first fully successful full-range electrostatics, and also among the most respected, was produced in 1957: the Quad Electrostatic Loudspeaker (Quad ESL, later ESL-57) from Quad Electroacoustics, of Huntingdon, England. These were shaped somewhat like a home electric radiator curved slightly on the vertical axis. They were widely admired for their clarity and precision, but can be difficult to run while achieving low frequency bass output.
The Quad ESLs were designed by Peter Walker, founder of the company, and David Williamson. The first in the series was the ESL-57, influenced by U.S. Patent 1,983,377 developed by Edward W. Kellogg for General Electric in 1934.[4] It was introduced in 1955, put into commercial production in 1957, and discontinued only in 1985. In 1981, Quad introduced the ESL-63 as a successor to the ESL-57. It attempted to address both the deficiency in bass reproduction of the ESL-57 and its extreme directionality at high frequencies. The latter goal is achieved by splitting the stators into eight concentric rings, each fed with a slight time delay compared to the ring immediately inwards, thereby attempting to simulate a point source.
The ESL-63 remained in production until 1999, when Quad introduced the ESL-988 and the ESL-989— both of which the company still produces. Quad Electroacoustics introduced two new models in 2005, the smaller 2805 and the larger 2905. These models return to the slightly back-tilted stance of the original designs, albeit user-adjustable. Largely retaining the larger bass panels of the 98x models and concentric ring design of the ESL-63, the 2x05's feature heavier and far more rigid construction, and several electronic and transducer refinements.
Transparent Sound Solutions builds stand-alone electrostatic panels for OEM/ODM sales, with freestanding bass-modules as an option. Transparent Sound Solutions has several patents for design and industrial production of electrostatic panels. They use an inverted audio drive to the panels, compared to conventional electrostatic speakers. The standard drive method is to apply the high voltage bias to a high resistance coating on the inner diaphragm and apply the audio signal from a center tapped audio transformer to the low resistance outer stators. In the Transparent Sound Solutions design, the stators are high resistance and a complementary, meaning a plus and minus high voltage bias supply is connected to opposite stators. The diaphragm is then driven by the audio transformer. According to their white paper only half the required turns ratio is needed for the same output. This lowers the cost and size of the transformer and makes it an easier load to drive with an amplifier. A very thin (13.9 mm) panel design can be achieved, making them a good match for integration with LED-TV and lifestyle products..
Another full-range speaker that is out of production was the Canadian manufactured Dayton Wright XG-8 and the XG-10 from 1968 to the 1990s. They were distinctly different in design, by enclosing the panels in an airtight bin, containing sulfur hexafluoride, a gas used in high voltage arc suppression devices. This gas, even with a 50% dilution has a breakdown voltage compared to air of about 2.5 times. This allowed a much higher than usual polarization voltage of up to 16 kV, more than twice as high as any other electrostat. The advantages were higher sensitivity approaching that of conventional speakers and not requiring the insulation of the stators. The higher bias was not used to just increase the efficiency, but mainly to produce high output by increasing the gap between stator and driven diaphragm to almost 0.2 in. This coupled with an enormous step-up transformer allowed the audio voltages to also reach 16 kV for higher output levels than any other design then and now. With an amplifier of 500 W RMS per channel, sound pressure levels equal to conventional speaker could be produced. The bass response down to 40 Hz was achieved by several things. First the cell diaphragm coating had a resistance exceeding 1000 MΩ per square, requiring charge times of several days, but this increased the charge migration time to a few seconds, reducing the low frequency distortion. A large core transformer that did not saturate at 600 W at 20 Hz. Taking advantage of the advantage of the SF6 gas to provide heavier than air loading to the electrostatic cells to drive the front and rear external diaphragms being 1.6 times larger than the cells. In addition the propagation of sound is 2.3 times slower in the gas, giving the effect of a much longer acoustic path and so lowering the cancellation frequency of the speaker caused by the out of phase rear wave. In addition, the transformer primary inductance was resonated with a capacitor to produce a low Q circuit at about 45 Hz to lift the response. The earlier models (XG8) had 8 cells per speaker and the later models has 10 cells. The cells were not full range due to their width causing too much beaming at high frequencies and the attenuation of the front diaphragm. The high frequency driver was a Motorola piezoelectric tweeter crossing over at about 7 kHz which was then replaced with a Panasonic leaf tweeter in the later generation models, crossed over at 6 kHz. The piezoelectric tweeter was mounted inside the bin as it performed actually better with the gas than in air, but the leaf tweeter was mounted on the outside of the exterior diaphragm. Dayton Wright ,[5]
Other manufacturers currently producing electrostatic loudspeakers include Immersion from Australia;,[6] Solosound[7] in The Netherlands, King's Audio,[8] Panphonics[9] from Finland and Pune based Cadence Audio [10]
Audiostatic,[11] Sound Lab[12] exclusively build full-range electrostatic panels. The only active electrostatic loudspeaker currently in production is the Audiostatic DCA-5.
Innersound,[13] MartinLogan, Metrum Acoustics,[14] Sanders Sound Systems,[15] and Sound Lab,[12] build hybrid designs with conventional subwoofers.
Among electrostatic full-range speakers which are no longer made are the KLH 9, one of the earliest US full-range designs, although the bass dropped off rapidly below 70 Hz.[16] There were several Acoustat[17] models manufactured, and the Infinity Servo-Statik and its successors which used a dynamic subwoofer at low frequencies.