A STATEMENT OF POLICY
CORE TECHNOLOGY: THE DICK DIPOLE DRIVER
WHY THE DDD IS TRULY REVOLUTIONARY ?
OPTIMIZING THE TRANSMISSION LINE DRIVER
THE FREE LUNCH
CROSSOVER AND CABINET DESIGN
A STATEMENT OF POLICY
Almost everyone in the business of specialty audio professes to pursue the same goal, the achievement of the most accurate reproduction of sound possible, at least within certain price constraints. Everyone claims to strive to be the best, and many claim to have succeeded.
So given the fact of such assertions, how does one arrive at a just estimation of any given manufacturer’s true commitment to advancing the art of high fidelity and, coincidentally, of the desirability of patronizing that manufacturer’s products?
We urge the music lover in search of the best in sound reproduction to look beyond mission statements and even technical claims and focus instead on how a given company has defined itself in its relationship with its own products and with its customers.
We think it says a lot about us as a company that we spent almost the entire first half of our corporate existence doing fundamental research in transducer design and nothing else. Not market research, not research on how to attract venture capital or to frame a promotional message, but research on how to make a better loudspeaker that would ultimately provide a more rewarding listening experience. We also think it is highly significant that we have sought to establish an ongoing relationship with each of our thousands of customers by designing and building products that will perform magnificently literally for decades on end.
German Physiks loudspeakers, unlike conventional audio components, are in no sense consumer items. They are not designed to be consumed, they are not meant to be disposable, rather are intended to provide outstanding performance literally for the life of the customer. In fact, few German Physiks loudspeakers ever change hands, and those that do command among the highest resale values in the industry. In a word, they are built for permanence.
In the following sections we explain how we developed our key technical innovations and what they mean to the reproduction of music in the home, your home. And we invite you to contact your nearest authorized German Physiks representative and experience those innovations first hand.
In 1985 Peter Dick, engineer, mathematician, and information technologies expert, was a very frustrated man. Dick, who had no professional involvement in audio engineering at the time, had yet become fascinated with certain fundamental problems in transducer behavior, and after years of mathematical modeling and physical experimentation, had created a design that he believed decisively surpassed the then state of the art. By 1985 Dick had succeeded in developing an extremely impressive working prototype based upon his innovative design concepts, along with a complete theoretical model of the device’s operation, and, he expected, at the very least, to elicit some degree of interest within the loudspeaker manufacturing industry. But of the many driver manufacturers he contacted throughout Northern Europe, most responded with condescension or disinterest, and none with enthusiasm.
Nearing the end of his manufacturer’s list, Dick finally called upon a medium sized German company named Manhattan Acustik, which he had no reason initially to believe would show any more interest than had the others. But Manhattan’s owner, Holger Mueller, was intrigued from the onset. The prototype which Peter Dick had developed was quite obviously a derivative of a design for which Mueller already had the highest regard, the famous Walsh Driver, invented by maverick American transformer engineer, Lincoln Walsh. Mueller himself owned Ohm F loudspeakers utilizing an early version of the Walsh driver, and had always felt that the design had enormous untapped potential. And, as he examined the crude looking prototype Dick had brought him, and pored over the detailed design notes, he realized that much of that potential had now been achieved.
In due course Mueller agreed to license the design, and thus German Physiks was born and dedicated to the purpose of turning Dick’s prototype into a real commercial product. And for seven years thereafter the nascent company would pursue that task, patiently extracting every increment of performance of which the design was capable, and producing no revenue whatsoever, instead deriving all of its operating capital from the parent company of Manhattan Acoustics.
During that seven year period the Dick Dipole Driver, as the new design was called, was refined and re-refined. Not satisfied with superb laboratory measurements and the unbridled enthusiasm of his listening panels, Mueller initiated an exhaustive life testing program to ensure the absolute reliability of the driver under the most abusive conditions of use. At the same time, aesthetics were not neglected, for Mueller believed that the industrial design of the driver must strongly express the company’s commitment to innovation, through-engineering, and overall elegance of visual form.
At last in 1992 the first production run was completed, and the first assembled speakers were introduced to the press and to the retail establishment. Acceptance of the merits of the new design was immediate and overwhelming on the part of both reviewers and the buying public, and the DDD and the loudspeakers incorporating it quickly won accolades as the most advanced sound reproducers in the world. And the praise has never ceased. Today German Physiks enjoys a position of pre-eminence at the topmost stratum of the audio marketplace in both Europe and Asia.
Although German Physiks cabinets and crossovers are second to none, it is our drivers that really set us apart and enable our speakers to achieve their unmatched fidelity. In this section we’ll discuss what makes them unique and uniquely accurate.
All loudspeaker drivers including the DDD may be placed within a few fundamental categories reflecting overall design characteristics, and those categories in turn can be combined into two groupings which are yet more fundamental, and, which as it happens, are not mutually exclusive. The first cardinal grouping is based upon the means of electromechanical transduction, i.e. how electrical energy representing the AC audio waveform is converted into mechanical energy, while the second organizes loudspeaker types according to how that mechanical energy is subsequently converted into acoustical energy.
The first class breaks down into three basic designs: electromagnetic, also known as electrodynamic, where a simple reciprocating motor drives the diaphragm or sound producing element; electrostatic whereby air motion is produced by varying the electrical charge of a diaphragm seated between two electrodes; and finally the piezoelectric type which employs certain materials which flex in one dimension when an electrical potential is applied across them. While other principles of transduction have on occasion been invoked by designers, such as magnetostriction and corona discharge modulation, no such variant approach has achieved commercial significance, and thus these three subcategories may be said to encompass almost every loudspeaker in existence at this time.
Well over 95% of speakers sold today are members of the first class, the electromagnetic, that is, they are essentially motor driver. The DDD is motor driven as well, its one point in common with conventional designs.
Of these three subclasses, both the electromagnetic and the electrostatic have achieved high levels of linearity while the piezoelectric has generally been confined to applications where superior fidelity is not a requirement. Of the two high fidelity types, the electromagnetic is the transducer of choice in our view. True, electrostatics are held in high esteem by many audiophiles, and deservedly so on the basis of their sound quality at moderate output levels, but they suffer from inherent limitations in excursion and output capabilities, and also from low electrical efficiency and poor reliability. Moreover, they are difficult to match to conventional amplifiers. Like all serious researchers, we have investigated them thoroughly. Regretfully, we have concluded that the electrostatic type is inherently impractical and unlikely to progress much beyond its current state of development.
The second fundamental grouping, that relating to the mechanical design of the diaphragm itself, includes the following: mass loaded pistons represented by ordinary cone and dome drivers; film and leaf transducers which include all electrostatics as well as ribbon type dynamic loudspeakers; flat panel speakers such as the NXT and BEST; the Heil air motion transformer which is essentially a class with only one member; and finally transmission line drivers, chiefly represented by the Jordan Module, the Manger, the Walsh Driver, and our own DDD.
The mass loaded piston class includes nearly all conventional cones and domes made today or in the past. The term itself refers to the dominant effect of mass on the acoustical output of the driver, and to the theoretical model of pistonic motion to which such drivers conform to a greater or lesser degree.
In such designs, mass reactance will be the main component in the complex acoustical impedance of the diaphragm throughout most of the useful frequency range of the driver, hence the term, mass loaded. The mass in turn is loaded by the compliance or springiness of the driver’s suspension, and the two together, mass and compliance, form a resonant system like a weight suspended from a spring. Such a system tends to oscillate around a single frequency when excited, and, predictably, a large part of conventional loudspeaker design is taken up with damping such oscillations.
Unfortunately, a resonant system, even a damped resonant system, is poorly suited to sound reproduction. Audible sound spans twelve octaves, while a strongly resonant system is mechanically efficient only in the vicinity of its resonant frequency. Such systems are necessarily bandwidth limited due to such frequency dependent efficiency, while transient response is inevitably degraded due to inertia imposed by mass and because of the energy stored in the compliance and returned subsequently.
Such a driver, unless it is loaded into a horn or acoustic lens, will also suffer from a nonconstant directivity pattern, with dispersion normally narrowing with ascending frequency. This characteristic, perhaps more than any other, will cause a loudspeaker to sound musically unnatural since acoustical musical instruments almost never radiate sound in this manner.
The normal strategy for dealing with the limitations of mass loaded pistons is to use two or three per speaker system and assign them to narrow frequency ranges around their respective resonant frequencies. Many fine speaker systems have in fact been designed according to this pattern, but the approach, we believe, is fundamentally flawed, and nearly always results in compromised transient response, ragged directivity patterns, and an overall sense of individual drivers imperfectly integrated.
The second major grouping of loudspeakers in terms of mechanical design and behavior is the tympanic or membrane variety, including dynamic ribbon and leaf types as well as electrostatics. All of these characteristically use low mass, large surface area diaphragms stretched over a frame; in other words, driver and suspension are one and the same. In such designs the resistance of air dominates the acoustical impedance of the diaphragm except at the lowest frequencies, and consequently transient response and frequency range can be excellent. But because these designs normally lack high power handling capabilities and are incapable of large excursions, they are best confined to the higher frequency ranges and indeed are thoroughly impractical in bass applications. Moreover, they typically dictate a line source configuration which makes them difficult to deploy in domestic listening spaces. They also tend to be difficult to integrate with mass loaded pistons due to their very different acoustical charactertistics, and so hybrid systems incorporating them generally provide disappointing results.
The Heil Air Motion transformer, our class of one, is worth an essay in itself, and is an extremely ingenious design with high output, high efficiency, wide bandwidth, good impulse response, and low distortion. Its sole real drawback is its low frequency limit—not much under 1kHz—and the consequent necessity of mating it with a conventional woofer. Sadly, most attempts at doing so have resulted in embarrassing mismatches.
The transmission line type has commonly employed a steep, straight-sided cone and a fairly conventional voice coil and magnet assembly. But where it differs from an ordinary mass loaded cone is that the diaphragm is securely anchored at its mouth and flexed by the motions of the voice coil rather than pushed to and fro. Sound propagation is normal to the slope of the cone rather than parallel to the path of the voice coil in the gap as is the case with a mass loaded cone.
The diaphragm itself ideally has an extremely high stiffness to mass ratio, but because the diaphragm is extremely thin and its moving mass is extraordinarily low, so perforce is the bending resistance. Consequently, the diaphragm will be excited into bending modes quite easily; that is, it will not behave as a perfect piston at any frequency.
Without going into the physics of traverse wave propagation across a plate structure—for that, essentially, is what the diaphragm is in this design—we can say that when the plate is bent by an actuator, the actuator itself—in this case the voice coil—sees only a very small increment of mass from the diaphragm. Rather than being mass loaded, it is loaded instead by the radiation resistance of the air load on the diaphragm and secondarily by the stiffness of the diaphragm.
In layman’s terms, the voice coil is exciting shock waves across the surface of the diaphragm which in turn excite motions in the air. As distinct from a conventional cone, there is almost no mechanical inertia to overcome, and thus there is a very direct translation of electron motion into the motion of air molecules in the listening space.
In a real sense, the acoustical behavior of the system is much closer to that of an electrostatic membrane speaker than to a mass loaded cone, to which the transmission line driver bears a misleading external resemblance. The moving mass of our own DDD is under two grams, less than that of most tweeters, and yet its ability to displace air is roughly equivalent to that of a 6 1/2" woofer. So while it shares the cone shaped diaphragm of the latter, its behavior could not be more different.
The combination of high displacement, low mass, and high acceleration allows the driver to operate linearly over a very wide frequency range, nearly the whole of the audible spectrum, as it turns out, and to achieve excellent impulse response and flat phase response into the bargain. At a stroke most of the limitations of the conventional cone are eliminated.
Mass producing a reliable transmission line driver happens not to be very easy, however, nor is the design process involved in extracting the fullest measure of performance from this type of loudspeaker. The first two transmission line drivers to reach the market, the Jordan Module and the Walsh driver, were both notoriously fragile and electrically inefficient as well as manifesting audible problems at the frequency extremes. When we revisited the design in the late nineteen eighties, we were determined to do better.
OPTIMIZING THE TRANSMISSION LINE DRIVER
Increasing the sensitivity of the transmission line driver entailed augmenting magnet strength and refining the magnet architecture. Fortunately, we had at hand design tools and magnetic materials not available to the pioneering designers of transmission line drivers, and we were able to arrive at a suitable design fairly early in the engineering process, one that overcame all of the previous limitations of the magnetic circuit endemic to the design.
The current DDD utilizes an extremely powerful rare earth magnet of our own design and manufacture, and an underhung voice. Sensitivity is on a par with that of conventional cone drivers of similar dimensions, and the linearity of the magnetic circuit—a grossly underappreciated factor in the overall performance of any dynamic driver—is among the best in the industry. As a matter of interest, magnetic strength in the gap is approximately 2 million gauss.
Since, in the interest of limiting the moving mass, the amount of conductive material in the gap must also be minimized, power handling obviously becomes an issue in this type of design, and was indeed a critical limitation in earlier examples. Today we achieve power handling on a par with that of conventional woofers by using a very wide voice coil which dissipates heat easily, and by completely enclosing it within the magnetic structure which then serves as a heat sink for the voice coil.
The other design problems associated with the transmission line driver are subtler but no less real and required far more extensive research to solve. These have to do with the vibratory behavior of the cone when excited.
In any driver of this type, one wants the output of the cone confined to precisely the same number of wave cycles represented by the electrical signal—in other words, one doesn’t want the cone to exhibit „ringing“, a type of behavior in which the cone continues to move after the exciting impulse has spent itself. Essentially perfect control of the diaphragm and an absence of ringing is easily enough achieved when the wavelength of the frequency propagated down the cone is greater than the dimensions of the cone itself, but is difficult when the wavelength is shorter, since the full wave then becomes subject to reflection from the boundary of the cone, and the reflection will in turn engender rereflections as the traveling wave slowly loses energy over the course of several wave cycles. Imagine small ripples in a pool and how they are reflected back from its edge in a recurring pattern. The behavior of a rippling diaphragm is precisely the same, that is, motion tends to persist for a considerable duration which ultimately has the effect of obscuring information in the recording.
Pioneering researchers such as Lincoln Walsh and Ted Jordan were certainly aware of this problem but lacked the analytical tools and materials technology to solve it effectively. They responded with cut and try solutions which were at best partially effective.
Our own approach is two fold. We have engineered our diaphragm to exhibit increasing velocity of propagation with increasing frequency, and we have devised effective means of damping the residual ringing in the cone which cannot be eliminated by the first method.
Velocity of propagation is another way of saying speed of sound. Titanium, our material of choice for the diaphragm, exhibits a very high velocity of propagation in all instances, but especially when it is worked into a very steep thin-walled cone, a shape which maximizes its inherently high stiffness to mass ratio. When such conditions have been met, peak velocity will reach several thousand miles per hour, many times the speed of sound in air.
Interestingly, when the dimensions of the cone are just right and the titanium foil is treated with certain applications, the speed of sound through the diaphragm will actually increase with frequency, a highly desirable state of affairs as it happens. The practical consequence of this frequency dependent velocity characteristic is that the wavelength remains relatively constant regardless of frequency since by increasing its velocity a wave can be made to travel farther in a single cycle, far enough to compensate for the naturally decreasing duration of the wave cycle with ascending frequency.
Why is a constant wavelength independent of frequency so desirable? For two reasons. First of all, because the wavelength tends to remain greater than the height of the cone for several octaves, which in turn prevents reflections and the resultant phase anomalies, and maintains the entire wave form as a pressure node in direct and intimate communication with the voice coil. In the DDD almost half of all audible frequencies are reproduced as fractional wavelengths, and fractional wave lengths, as wave theory teaches us, cannot be reflected.
The second benefit is less obvious but scarcely less important. Because in the DDD wavelengths are approximately equal in length across a wide frequency range, the movement of the diaphragm engenders almost no Doppler distortion, the bane of conventional loudspeakers. The high frequencies do not ride on top of the lower ones as they would were they shorter in wavelength, and a low frequency can act upon at most a fraction of a wave cycle of a higher frequently. Simply put, there is no mechanism for generating Doppler.
The total bandwidth over which this phenomenon of frequency dependent velocity of sound will manifest itself depends to a great extent on the thickness of the titanium foil. All things being equal, the thinner the foil, the wider the bandwidth in which the condition obtains. The foil we use is on the order of one millimeter in thickness, very thin indeed. We have experimented with still thinner sheets, but have found the resulting structures unduly subject to accidental permanent deformation, and have concluded that a one millimeter thickness represents the best compromise.
At its current dimensions, the DDD reaches what is known as its dipole frequency at a little over 7kHz. At this point the increase in propagation velocity no longer matches the increase in frequency, and the wavelength becomes progressively shorter than the cone itself thereafter.
Once beyond the dipole frequency, damping mechanisms must be invoked to control the motions of the cone. The precise nature of these mechanisms constitutes a trade secret and will not be divulged here. Suffice it to say that they control ringing very effectively and allow the DDD to operate up to 19kHz, within a semitone of the figure conventionally cited as the high frequency limit for human hearing.
A transmission line driver used in a Walsh configuration, i.e. mounted upside down atop a cabinet, possesses an added advantage that is arguably just as great as the unrivaled linearity of the design. That advantage is a nearly ideal point source omnidirectional radiation pattern.
The DDD propagates sound in a perfectly uniform spherical pattern just as do musical instruments. Frequency response and phase response are absolutely uniform from all listening angles which is never the case with two and three-way cone and dome loudspeakers nor with dipole electrostatics and ribbons.
As with conventional cone drivers, the DDD requires an enclosure to contain the back wave. Except in the case of the fully horn-loaded Unicorn, the DDD is loaded into a sealed chamber in all models. Again excepting the Unicorn, all of our systems are augmented with integral subwoofers.
Listening evaluations and instrument test have both revealed that „cabinet talk,“ the re-radiation of sound through the walls of the cabinet is one of the worst offenders in detracting from a sense of realism in sound reproduction and a lowering in the perceived resolution of a speaker system. To quell cabinet talk we damp the walls of our cabinets very effectively by means of Hawaphon, an extensional damping material consisting of a polystyrene quilt containing numerous small pouches of fine steel shot. Originally developed as an antisurveillance measure for use in military and governmental offices, Hawaphon achieves broadband attenuation of structure-borne sound of more than 50dB, a truly remarkable figure. Loudspeaker enclosure panels treated with Hawaphon generate almost no spurious output.
As an option we offer an ultimate level of cabinet construction based upon dual laminations of resin impregnated bidirectional woven carbon fiber bonded to an MDF core. Interior applications of Hawaphon are used in addition to carbon fiber reinforcement with resulting structures achieving a new level of cabinet quieting for the industry. Carbon fiber cabinets are significantly more expensive to build, but the augmentation in performance is not subtle. For those for whom even the slightest compromise in performance is anathema, we recommend carbon unreservedly.
German Physiks passive crossovers are the product of extensive computer modeling based upon our own design software, which, incidentally, is rapidly becoming the reference for the industry. They utilize only the finest components, and raw parts cost for some of them exceeds the retail price of typical moderately priced branded loudspeakers. We also make a dedicated electronic crossover for use with our component subwoofers. Construction is true dual monaural including the provision of a separate power transformer for each channel. Meticulously engineered and massively overbuilt, our active crossover exceeds the size of most power amplifiers, and engenders none of the veiling associated with lesser examples of the art.
All German Physiks loudspeakers utilize the same DDD bending wave driver, and all are capable of a degree of resolution significantly exceeding that of conventional designs. All feature exceptionally well damped, exquisitely finished cabinets, and superior passive electrical components. Where our various models differ is in the number of DDDs used per cabinet, the selection of subwoofer drivers, and in the complexity of the crossover networks. Larger systems achieve higher output levels and more extended bass response due to the sheer amount of driven surface and due to the larger enclosed air volumes. Because of the high inherent manufacturing cost of the drivers we utilize, scaling up our designs is not inexpensive, but for those who demand the best, nothing else is adequate.
We hope this explanation of our design philosophy assists you in choosing a high performance speaker system, and urge you to visit your nearest authorized dealer. Don’t deny yourself the experience of music over a German Physiks loudspeaker system.