EP2572515A1 - Split-membrane 3d stereomicrophone - Google Patents

Abstract

Split-membrane 3D stereomicrophone for converting the sonic actions of the surrounding air to two electrical voltages, consisting of a microphone capsule in the shape of a hollow body having at least one opening that is covered with one membrane each as the electrode, said membrane consisting of a flexible and electroconductive foil and being fastened in an electrically insulated fashion and spaced apart from at least one metal plate as the counter electrode which is fastened inside the microphone capsule, the metal plate being subdivided into a left-hand metal plate as the left-hand counter electrode and a right-hand metal plate as the right-hand counter electrode by a gap, said electrodes being electrically insulated from each other and having a respective electrical connection of their own.

Description

 3D stereo microphone gap

 The invention relates to a 3D stereo microphone according to the preamble of claim 1.

For recording and reproducing noises and sounds, electromechanical transducers have been known for over a century, which convert micromotion movements of air molecules into electrical signals that are processed and stored. At a selectable point in time, the stored signals can be retrieved and converted back into sound - that is, in movements of air molecules - via other electromechanical transducers, which are referred to as loudspeakers.

Since March 2006, DEGA recommendation 101: Acoustic waves and fields has been in existence by DEUTSCHES GESELLSCHAFT FÜR AKUSTIK eV, Voltastr.5, Geb. 10-6, 13355 Berlin. It can be read on page A10 on the acoustics: "If, however, complex perceptions are evaluated, for example, in the quality assessment of speakers, subjective tests can not be waived." Related to human perception, are the numerous, currently The term "information" is preferred in this patent application as a generic term for the audio-acoustic representation, which according to its Latin origin in transcription, the "in-form-making" of The fourfold complexity of the human perception of "location and shape and sound and environment" contained therein replaces the terms used in most disciplines, such as signal, message, pattern, shape, bang, noise, sounds, sounds, and spatial / binaural Listen. The original goal is to reproduce as accurately as possible at the ear of the hearing human being the same shape movements of the air components that would have reached the ear if the hearer personally and directly heard the respective noise. Ideally, the information about all individual movements in the air must be stored and reproduced completely and accurately.

In the current state of the art, however, only a more or less large proportion of the movement information without falsification returns in practice on the way from the first conversion in the microphone via the processing and storage of the electrical sound information and the second conversion in the loudspeaker air movements. It is interesting to note that the hearing-adapted audio coding method is described in the literature as "not preserving waveform" (Thomas Sporer, Faculty of Technology, University of Erlangen-Nuremberg-Ilmenau, Dissertation 1998, p.8).

In order to show that conventional loudspeaker systems and the previously known microphones can transmit only simple waveforms, a comparatively very detailed description of the physical fundamentals and the state of the art achieved so far is required. Only on this basis is the aim of this invention recognizable and comprehensible, in which the vecto-bright information in directional form movement are guaranteed, so that the listener to the sounds and sounds and their sound location and their identity form precisely, and thus lifelike in 3D can experience.

In the current state of the art, it is almost exclusively regarded as the task of transducers, periodic oscillations Both in the conversion of sound into electrical voltages - so a microphone - and in the conversion of electrical voltages into sound - so in a speaker. Therefore, all waveform details in sound are almost always represented as a superimposition of waves or oscillations with different amplitudes and frequencies.

However, it is neglected that sound arises even at the slightest movement of an air molecule. Accordingly, the latest findings show that this focused only on waves, oscillations and other periodic forms of movement approach is too narrow, because actually the movement of a single air molecule is already sound.

It is therefore a very important finding that fundamentally expands the basis of the previous acoustics, that a non-periodic form of motion of the air molecules is information in the sound and must be taken into account when designing electromechanical transducers. Therefore, for the highest demands on transducers, it makes sense to test and optimize their properties and abilities with regard to nonperiodic and periodic forms of motion.

For the characterization of sound, the respective perimeters of the following three concepts must be examined more closely, namely "rest" and "momentum" and "vibration or wave." So far, with regard to sound, only the vibration has been regarded as a stationary, bodily and periodic form of motion.

This horizon is clearly too narrow and must be based on nonperiodic forms of movement as well as the contrast to these movement forms. The "rest" is initially defined by the quasi-static pressure of the ambient air, ie 1013 hectopascals.

The "brown motion" of the air molecules taking place therein is, strictly speaking, already a form of sound, which is also called "background noise". It is well-known that "well-being", ie the normal state of hearing, includes the perception of this noise floor.Experiments have shown that persons who have been kept away from background noise for a longer period of time experience this as a burden.Humidity thus clearly detects the background noise true, but classifies it in the evaluation in the brain as a rectilinear motion-form noise of an inconspicuous environment and thus not as sound.

Only when a sound impulse as a curve shape deviation stands out from this background noise, the curve shape is evaluated as currently new Schali information. This results in a recognition "approach" for the fourfold Schali information from place, form, sound and environment in total sound perception.This, in English also called "onset" mechanical impulse change in the sound exceeds the ground noise, that is also a basic hearing threshold, or a monitor zero line. This relationship is a clear demonstration of the hitherto significantly underestimated abilities of the human hearing organ and in particular of their evaluation in the brain in its temporally four-layer complex evaluation in chronological order, as Figure 1 shows.

The example above shows that astonishingly low air movements are already absorbed by the human ear and get ranked. Therefore, for the definition of a sound conversion of the highest possible quality, that is to say a sound conversion which in the ideal case is no longer perceptible as such by the human ear, the primary physical cause of the human perception of sound must again be considered.

This inevitably leads to Newton's laws as the fundamental law of classical mechanics: The notion of "rest" is related to Newton's law of inertia, which states: "All bodies remain in the state of rest or state of inertia in the absence of forces uniform straight-line movement. " It is interesting that the law of inertia is already of utmost importance for the sound information and the construction of sound transducers. Because in the cochlea in the inner ear of the human is a straight line measurement, which are formed by the bipolar inner hair cells, the IHZ, where all form deviations in the direction and amplitude as vector magnitudes.

The movement of the mass, which contains a body, is called in the German language "impulse", in other languages also as "momentum". Just like the speed of this movement, the momentum is a vector quantity, so besides an amount, it also has a direction.

In Newton's mechanics, the momentum of a particle is the product

from its mass m and its speed :

Definition: pulse state

Pulse and velocity are vectors with a certain amount and a certain direction.

In contrast to the pulse state, however, the change in momentum is of decisive importance in the development of a sound movement, because only then is the sounding body deformed and / or accelerated in its mass volume because forces exert their effect on the body in two ways.

Important for the sound movement change is the dynamic force law or action law of Newton. It is also called the "Basic Law of Dynamics" and forms the basis for establishing the equations of motion for numerous systems of mechanics, including for sound movements in the air:

"The change in the motion of a mass is proportional to the action of the moving force, and it is in the direction of the straight line upon which that force acts."

Or expressed in a formula: The force F, which causes a change of motion, is proportional to the momentum change over time change with constant mass m:

with the definition of the force unit F = m a

or force = mass ^. acceleration

An impulse of 1 N s causes a momentum change of 1 kg on a body ^. m / s caused.

Newton's formula clearly states that labeled quantities are vectors, that is, quantities directed in space. This also applies to sound movements. In order to evaluate the loss of information in conventional transducers, it must also be fundamentally clarified what information the human ear actually receives when listening and how it evaluates it.It is part of the general knowledge that the human ear different sounds from a mixture sinusoidal sound waves with different frequencies and amplitudes, as produced by musical instruments, for example, so that the sound of an oboe is differentiated from that of a violin when heard.

There is a growing realization that human hearing is by no means limited to the detection of stationary, periodic vibrations of the air. A no less important task of hearing is, on the contrary, the recognition of short-term, dynamic air pressure changes, so-called transients. They are caused by noises such as the cracking of a branch or the firing of a bullet.

It is now generally accepted that humans also have a recognition mechanism for such sounds. It is noteworthy that this recognition responds much faster and leads to the perception of a "heard information" much earlier than is the case with sinusoidal oscillations.

Already tiny changes in the static air pressure distribution, which arise eg due to the force effect of a breaking branch or an impacting object, are not only immediately recognized by humans, but can also be assessed according to their direction after a - otherwise imperceptible short time. The explanation for this is an alarm mechanism developed over many developmental stages of the hearing and brain. The origins of human hearing are alarm messages that serve to locate a danger and to determine its character. In the evolution of primates those individuals have prevailed that could detect their "predator" so early and localize so early that they successfully escaped him.

Therefore, the brain of modern man as an "artifice for identifying the direction of the initial sound that the level of the primary noise pulse on its way to the brain of specialized nerve cells up to thirty times amplified. During this time, the impulse rises clearly from the normal level and returns a few milliseconds thereafter to the normal state, in which the ear is given priority to the perception of periodic signals such as, for example Tones is focused.

Meanwhile, in the brain, the signals of the left and right ears can be compared and evaluated so that the direction of the noise pulse can be located. This ability is also known as binaural hearing.

For all sound signals that do not strike the human head head-on from the front, but come from other directions, so-called interaural time differences (ITD) and so-called interaural level differences (ILD) arise.

Runtime differences can be evaluated by the human ear already from a size of 10μs for directional localization, which corresponds to a localization sharpness of about one degree. To to a transit time difference of 630μs, the localization increases approximately proportional to the transit time difference. A transit time difference of 630μs corresponds to a path difference of the sound of 21, 5cm. This size is also called "Hornbostel-Wertheimer constant" and corresponds to the average distance between the two

Ears on the human head.

With relatively slowly increasing pulses, the brain evaluates primarily the phase differences between the signals of the two ears and determines from this the interaural transit time difference (ITD).

For very fast rising pulses, the localization is based on the evaluation of interaural level differences ILD, as well as on the evaluation of interaural group delay differences, ie the propagation time differences of the signal envelopes.

In an intervening region of average steepness, the range of effect of the involved effects overlaps. With increasing steepness of the pulse, the angular range in which interaural phase differences can be evaluated becomes smaller and smaller. For this, the size of the interaural level differences increases.

F. Pfander also writes "Das Knalltrauma", Springer 1975, Berlin and "Das Schalltrauma", BMdV Bonn 1994. In

Based on this, FIG. 1 shows the course of the air pressure as a function of time at the ear of a listener when a dynamic air pressure change is generated at a certain distance from it by a single, rectangular force, hereinafter also referred to as "dynamic pressure change" , Around- In the vernacular, such a dynamic pressure change is also called "bang".

It has its origin in the so-called onset time to, from which the air pressure initially rises within the time interval to-ti up to the highest pressure value Pm. This is followed by an aperiodic pressure equalization, during which the air pressure drops below the atmospheric air pressure Pa from the time t2 and has risen again to the atmospheric air pressure Pa at the time tv.

The course of this pressure change can be divided into four different characteristic time zones, which are characterized by different phenomenological effects in humans. During the first pressure change range in the time interval to - ti, the force acting in onset time to accelerates the relevant air masses.

This acceleration can also be described as a temporal change of the speed. Depending on the size of the maximum pressure value Pm, the variable velocity air molecules cause the impression of a "pop" or "crackling" in the human ear.

In addition to the information that there is a sound at all, the human ear is guided by the difference in transit time of the sound

Sound between the two ears sound information about the spatial origin of the sound, by determining the direction from which it comes. For this purpose, it is already possible within the time interval to - ti, since, as already mentioned above, the maximum detectable transit time difference of the sound in the evaluation over both

Ears may be about 630 με if an average ear stand of 21 cm, a sound velocity of 343 m / s and an exactly lateral incidence of sound in the direction of an - imaginary - connecting line between the two ears can be assumed.

After measuring the direction of the sound, the brain concentrates in a second time zone ti - tv on the recognition of the cause of the sound impulse. Typical for this time zone is the compensation of the maximum achieved air pressure value Pm. This pressure equalization produces another sound, which resembles a so-called "evocative keynote".

The term "root" refers to the music that deals primarily with "tones", so with periodic vibrations. These vibrations are almost never exactly sinusoidal in practice, but can be broken down into a sum of numerous sinusoids, which is referred to as Fourier analysis.

The resulting vibration at the lowest frequency - that is, at the longest period - is called the "fundamental tone." All other higher-frequency vibrations are so-called "overtones." The "root" determines the perceived pitch, while the resonating "harmonics" determine the timbre of the tone. If the overtones are rather low, one speaks of a "fundamental" sound.

In the second time zone ti - tv, the type of fundamental sound recorded by the hearing organs depends, inter alia, on the temporally transported energy of the jump excitation. Experiments have produced the surprising result that in the second time zone ti - tv the pressure change in the form of a "decaying fundamental" in the hearing areas of the brain causes almost no pitch sensation, although a pressure change of the same course in the fourth time zone - about 30 Therefore, the two time zones of to - tv are also referred to as "masking time". Although the pressure change in the second time zone ti - tv is periodic and not abrupt, no pitch sensation is possible before the time tv is reached. Instead, the brain is primarily focused on detecting the nature of the sound.

Adults may already be in this second time zone e.g. to distinguish a trumpet from a violin. This is particularly noteworthy because with a continuous tone these two types of musical instruments have very little difference in their sound. During this time, the comparison with learned noise patterns takes precedence in the brain.

The duration of the masking time is depending on the suggestion z. 10 ms to 22 ms, which corresponds to a sinusoidal or cosinusoidal onset excitation of a number of approximately 9 to 22 periods of a 1 kHz sine wave. According to the empirical findings of the inventor, only the location and type of a sound source are perceptible before the end of this time interval to - tv.

The third time zone is the so-called blur area around the time tv. In this time zone, the brain is no longer limited to comparisons with noise patterns in the evaluation of pressure changes, but it sets in the pitch sensation. The Sharpness range begins - depending on the individual about 10ms after the onset time to and ends at the latest about 30 ms after to.

Only in the following, fourth time zone, that is - about 30 ms after to - the brain analyzes the pressure changes primarily on the frequencies contained therein and their share of the frequency spectrum. Only after this point in time does the brain begin to realize what has often been understood as "hearing." In fact, however, listening is from the first time zone, which begins with the onset at time to and almost until time H The second part of the "hearing" is the second time zone ti - tv, which is mainly used for comparison with learned sound patterns. Only in the third time zone, the blur range from about 10ms to about 30ms to to, these processes are completed and the fourth time zone begins with the recognition of pitches and timbres. Of these four time zones of hearing, the first three are still often misunderstood today. One explanation for this is that during these first three time zones, pitches and timbres are almost unaware, and thus are "hidden" in the evaluation in the brain.The frequency analysis of sounds and sounds, which is still predominantly considered today, only sets after the sound triggered Masking time tv, which begins with the onset time to.

In the following, another important effect in "hearing" is described, which makes it clear that the human ear is over

Has abilities that are about perceiving even go far beyond periodic movements. It is known that the ear assigns a pitch to a harmonic sound whose frequency corresponds to the already mentioned fundamental tone. When breaking down a sound into a multitude of sinusoidal vibrations, also called partials, the so-called fundamental tone is the lowest frequency sub-tone.

That when listening to all the partials that the sound contains, just the lowest is made the lead tone, which then determines the whole sound by its pitch, could be explained by the fact that in most sounds, the fundamental tone of that part of the highest amplitude is. And, as a rule, the fundamental tone actually dominates among the partials.

Therefore, until a few decades ago, it was believed that the fundamental tone determines the pitch of the entire sound due to its spectral dominance. The experiment described below shows very impressively that this is not the case: when an electronically generated, sawtooth-shaped sound is emitted through a loudspeaker, the ear feels the pitch of this harmonic sound as expected at the frequency of the fundamental. And in fact here is the root note with the largest amplitude in the sound spectrum.

In the next step of the experiment, if the fundamental tone is filtered out electronically, a tone is created whose lowest part tone now has twice the frequency of the fundamental tone and the largest amplitude. Actually, in the first approach, this mutilated sound would sound one octave higher than the original, sawtooth sound. That's not the case. Although the sound loses much of its deep, full sound color, but its tone pitch has clearly remained the same, namely that of the - now no longer contained in the spectrum - fundamental.

It can therefore be assumed that the ear can determine the tone pitch from the totality of the partials. In the brain's hearing center, the experience is stored that a harmonic sound always consists of partials and that in a sound mostly the fundamental dominates, which is why basically the sound is assigned the pitch of the fundamental. If a sound appears in the ear, in which the basic tone is not dominated or even missing, it nevertheless reconstructs the familiar sound from the available information, namely from the remaining tones of tears. To a certain extent, the ear extrapolates the missing fundamental tone by deducing from the systematics of the existing harmonics the presence of a fundamental tone.

Other human sense organs also have similar capabilities, preferably for temporally stationary events. For example, the human brain has a good memory for image impressions, but less so for its temporal consequence. Accordingly, the "spectral apparatus of hearing", like any other spectral apparatus, is used to represent temporal changes as stationary events, so it could be that the effort of spectral analysis of the sound movements in the ear serves to obtain stationary information that can be read and stored.

When sound events strike the eardrum of the human ear, they are mechanically amplified by about 22 times via the hammer, anvil and stirrup as mechanical vibrations are transmitted to the cochlea. The cochlea consists - Rolled apart and simplified described - from three parallel, flexible hoses, also called Scala. The middle of the three tubes is the Scala media, which adjoins the Scala vestibuli on one side and the Scala tympani on the opposite side. At the end of the cochlea, the Scala vestibuli and the Scala tympani are connected and both are filled with the liquid perilymph.

The interface between the medial scala media and the scala vestibuli is the Reissner membrane and the connecting surface between the scala media and the scala tympani is the basilar membrane. The latter is narrow and stiff at the beginning of the cochlea and broad and soft at the end.

An incoming sound impulse is introduced from the stirrup via the oval window into the Scala vestibuli and propagates there in the liquid perilymph as a traveling wave. The pressure is passed through the Reissner membrane into the Scala media and from there through the basilar membrane into the Scala tympani.

Due to the characteristic of the basilar membrane, the speed of the traveling wave decreases over the length of the cochlea, and the deformation of the flexible basilar membrane increases. Thus, a deformation travels along the scala media, which at very steep increases in the sound impulse or correspondingly at very high frequencies has its maximum excursion already in the front part of the cochlea and at very slow impulse rises or at very low frequencies only in the rear part of the cochlea Maximum reached. Therefore, the length of the rolled-out cochlea of 31, 5 mm also corresponds to the lower limit frequency of the human perceived as a sound vibration of 16 Hz.

To transfer the information about this deformation are the hair cells, which are connected at one end to the Tektorial- or cover membrane and are connected at the other end on the basilar membrane with nerve cells. The tectal membrane is a narrow strip which extends parallel to the basilar membrane through the scapular media and has swung there on a longitudinal edge.

Also, the tubular scala media in the middle between the two tubes of the Scala vestibuli and the Scala tympani is filled with a liquid, namely the endolymph. The movements of the Scala vestibuli and the Scala tympani are therefore also passed on to the endolymph in the Scala media.

Through these movements, the tectal membrane and the basilar membrane also pivot against each other and thus excite the actual scarf isensoreri in the inner ear, namely the hair cells, which then transmit messengers that keep them in small vesicles, as information on nerve cells continue to the brain.

The hair cells transmit the information with the greatest temporal precision and show the highest transfer rates in the entire nervous system. Only since 2005 is it clear how the hair cells can achieve such high rates.

Claudius P.Griesinger from the Physiological Institute II of the Albert-Ludwigs-Universität Freiburg has collaborated with researchers from the University of Freiburg versity College in London found that, for example, the neurons of the cerebral cortex or the hippocampus release their messengers via vesicles, which they constantly recycle and refill, whereas the hair cells rely only about ten percent on the slow recycling of the vesicles. Most of their vesicles produce them continuously in their cell body, so there is always a large supply of messenger packages available for fast and accurate signal transmission.

With conventional nerve cells, only a maximum of about 20 vesicles per second can be released since recycling does not allow higher transmission rates. The Hair Cells In the ear, their supply of vesicles has increased the maximum possible release by orders of magnitude. As a result, they can transmit information at much higher rates of speed, which in the first instance allows a precise transmission of the temporal fine structure of acoustic information.

The hair cells are divided into two groups, namely the inner hair cells (IHZ) and the outer hair cells (ÄHZ). The inner hair cells in the Scala media recognize due to their mechanical linear alignment on the basilar membrane from otherwise uniform-rectilinear sound movements a curvature in the positive and negative directions as absolute physical time cause of an accelerated change in movement in the sound approach from the rest or its sonic extension ,

This vector sensor technology of the hair cells is made possible by the fact that, in contrast to the outer hair cells, they are not connected to the tectal membrane. Instead, they can float freely within the endolymph fluid of the cochlea. So far, they are represented in the literature as well as the outer hair cells as docked to the tectorial membrane.

The available analysis range is 31.5 mm cilia path length. This path length is traversed by an impulse impulse in about 5ms from the oval window to the heliocrotrema. The dispersion with its root from the velocity of the bending wave mechanism in the cochlea widens the ratio of the time range from 10 μs to 10 ms, ie 1: 1000, to 1: 31, 6 ms duration. Thus, the distance traveled is 31.6mm. In this straight line, approximately 3,200 bipolar hair cells (IHC) can be detected, which can detect the direction and the amplitude (maximum 30 dB) above the threshold of hearing as changing pulse formation in the sound as information (lat .: in-form-shaping) , One example is the so-called "beat tone" of a bell, the reason being that it is just a momentum impulse, without periodicity and of a momentary nature (Trendellenburg: Akustik, p. 59, Springer 1961).

For the complete pulse interval - as opposed to the "onset" time for location (localization) - the transverse vector circuit is available. Its perimeter is 2 x 5 ms x 3.14 (ττ), giving a run-time detection range of 31.4 ms. Thus, 31.4 ms are available for the sound-emitting body in its image as the source location and shape to the hearing (see FIG. 1). In this time span, the single image of an iconic representation is subconsciously presented in human perception as a hearing event, when the smallest point (3200: 31, 4 ms≈10 ms) for the central image capacity of 10 μs duration in its auditory acuity from the world of Technique of human hearing world can be made available. The new 3D stereo slit microphone technology is used for this purpose.

OHC-perception:

 The sensory cells docked on the tectorial membrane analyze, on the basis of their sinusoidal half-wave arrangement, also called V- or W-shaping, the periodic sound vibrations which, depending on their pitch, are mapped with their amplitude maximum at a specific location on the basilar membrane. In the three rows of the ÄHZ results in 30 dB in the amplitude maximum, which together with the IHZ up to 120 dB, the pain threshold leads (Wolf D. Keidel, Erlangen, physiology of hearing, Thieme Verlag Stuttgart 1975). Many ideas and models since Helmholtz have been published in the scientific literature.

This essential in sound transmission peculiarities of human hearing must also be taken into account by serving as a sound transducer speakers, because only then has a high-quality microphone, the electrical output signals has provided all the information contained in the motion pulses at the site of the formation of a sound or a sound are a benefit at all.

Since the quality of a chained transmission of a sound or sound pulse on movements of the air molecules, via a microphone that generates two electrical signals and two speakers, which again generate movements of air molecules from these signals to the human ear to be so high can, as the weakest link In this chain, must be discussed below on the quality of transmission of the speakers according to the current state of the art. For essential requirements in the transmission of sound are still largely ignored in the construction of most sound transducers today. For many manufacturers, even today a loudspeaker is considered optimal, which has a good uniform amplitude frequency response in the audible frequency range and can sufficiently reproduce continuous, quasi-static sinusoidal oscillations.

On the other hand, it is largely disregarded whether a curve corresponding to the dynamic pressure-time curve according to FIG. 1 can be achieved with a loudspeaker, which necessitates the complete transmission of all the information present in the time interval to-tv and thus for a signal which follows the signal dynamically exactly is.

So when it comes to the recording and playback of sounds and sounds, then the electromechanical transducers that convert the movements of air molecules into electrical signals as a microphone, as well as the - referred to as speakers - other electromechanical transducers, these signals again back to sound, that is, to convert again into movements of air molecules - ideally, all the information described above are transmitted.

In the current state of the art average speakers consist of a hollow cone, which is firmly clamped at its edge with a circumferential, resilient strip and is moved in its center by an electric drive back and forth.

In order to check to what extent such a conventional loudspeaker can accurately reproduce a sound impulse, it will be excitation is applied by its drive is connected to a DC voltage, whereby a dynamic pressure change is effected.

If the fidelity of the loudspeaker were to meet the aforementioned requirements, it would produce a curve for the change in pressure as a function of time, which would be identical to that of FIG. 1, during a dynamic, sudden activation of its drive.

However, quite different curves, which are shown by way of example in FIG. 2, result on the state of the art which is currently the most widely used for loudspeakers. FIG. 2 shows at the top left the rectangular electric input signal for driving the loudspeaker and, in addition, the theoretically ideal pressure change at a location in front of the sound generator. The bottom two rows show waveforms actually obtained and measured with a measuring microphone when six randomly picked-out conventional studio interception loudspeakers are used as sources to produce the dynamic pressure change of the air.

It can be seen that the curves within the first time zone t0-t1 according to FIG. 1 and the second time zone t1-tv are significantly falsified. In particular within the second time zone t1-tv, the theoretically ideal "decaying fundamental tone" actually to be reproduced by the jump excitation is superimposed by numerous very short noises generated by the respective loudspeaker which last about 0.1 milliseconds up to 20 milliseconds and from the loudspeaker diaphragm be developed at the beginning of each sound production. The loudspeaker diaphragm is a relatively inert mass which is pushed in one direction by the electric drive, opposing the deflection of the spring force of the diaphragm and its suspension until the mass comes to a standstill. Then the spring action accelerates the membrane again in the other direction, swinging beyond its original rest position. With this transient, the speaker settles to vibrations with a certain frequency.

Conventional speakers not only reproduce music at very high frequency, they also constantly produce a plethora of very short sounds, lasting from about 0.1 milliseconds to 20 milliseconds. The subdivision into high-frequency, mid-range and low-frequency loudspeakers also cause different natural oscillations in the conventional mass / spring systems of conventional transducers. In a conventional three-way box, the three very different transient noises of high-frequency, mid-range and low-frequency sounds can be heard more easily.

These distortions are one reason why, when conventional loudspeakers jump, the curve of the sound pressure over time deviates significantly from the ideal curve of a very faithful music and sound reproduction.

This effect is currently also discussed under the term "impulse fidelity." Pulse-fidelity is the ability of a loudspeaker to follow a pulse-shaped signal as far as possible without transient or decaying processes, and instead most loudspeakers produce state-of-the-art loudspeakers even vibrations with low, medium and high frequencies, which are caused among other things by partial vibrations on the membrane, a generally hard-hanging membrane and cavity resonances in the speaker and in the listening room.

When a conventional loudspeaker diaphragm is to emit a pulse, it triggers the movements of the loudspeaker diaphragm, which wave outwards. This will still radiate sound, although the impulse is long over. As a rule, the edge of the membrane is not terminated with the correct characteristic impedance so that the wave is reflected and the pulse is further extended.

This makes a perfect reproduction of the sound signals and the necessary pressure changes impossible, because in the first time zone t0 - t1 and in the second time zone t1 - tv a sound pulse in which the human ear and the auditory center of the brain actually the place and type of Detecting sound pulses, the sound pulse is not audible at all, but it only sound the operating noise of the transient loudspeaker. This directs the listener's attention to the speakers, their size, and their location.

After many years of research and development work by J. W. Manger, however, it has now been possible to create an electro-acoustic

To create a transducer which causes in a jumping excitation substantially corresponding to Fig. 1 curve. One result of this research and development is known under the name MSW bending wave transducer, as described, inter alia, in DE PS 18 15 694, DE 22 36 374 C3, DE 25 00 397 A1 and DE 27 25 346 C3. The principle of operation of this bending wave converter deviates dramatically from conventional transducers with conical membranes. Here, a soft, viscoelastic membrane is used for sound radiation, which has almost no restoring elasticity in the direction of their excitation. It generates bending or traveling waves, which migrate from the center to the edge and leak there without producing natural oscillations and / or standing waves. In addition, extremely short rise times in the range of up to 14 microseconds corresponding to the first time zone to-ti can be achieved. With such bending wave transducers, it is now possible to reproduce the curves of FIG. 1 almost unadulterated.

This is shown in FIG. 3 in a representation analogous to FIG. 2. The obtained air pressure change at the measuring location comes comparatively close to the ideal shown in the middle range. In particular, when using a bending wave transducer disturbing, caused by the speaker pressure changes almost completely missing, whereby the speaker as such is no longer recognized by the listeners and the original signal can be transmitted and felt virtually unadulterated.

The described differences between the two types of transducers are very noticeable especially in stereophonic reproductions with two loudspeakers. In the case of conventional loudspeakers, a stereo transmission is, despite the fundamental deficiencies mentioned, only reasonably trouble-free, as long as the listener is exactly in the axis between the two loudspeakers, ie. H. located in the so-called stereo center. If the listener is distanced only slightly from this axis, the disturbances become immediately extremely noticeable, since then the closer speaker and his / her / its are closer to the listener

Errors are subliminal and intuitively recognized. Being "intuitive" Information processing in the brain called, of which only the result, but not the processing process are perceived.

Biegewellenwandler according to the principle of J.W. Manger are, however, essentially free from natural oscillations, so that when removing a listener from the stereo center and a sophisticated sense of hearing does not change, since the nearer speaker is not perceived as disturbing due to lack of significant natural oscillations. This allows several people to enjoy the same content simultaneously. And especially in the approach of sound vibrations, the crucial part of every sound for the geometric location in the brain of the hearing person.

With the MSW converter, the music listener comes close to the illusion of sitting in front of the instrument itself, because the reproduction is virtually inaudible, since the MSW converter no longer emits transient noises. Instead of the sounds of the loudspeaker, the only thing that the ear hears is the transient effects of the musical instruments, which locate the instrument in the concert hall and reproduce its character image.

The decisive feature is the thin and flexible membrane, which consists of three layers. It does not store forces caused by drive or springback. Instead, the counteracting mass and spring forces cancel their storage components within the membrane itself. They are dissipated with the sound as heat. The membrane oscillates at a rise time of only 0.014 milliseconds to the signal frequency and holds this sound to the musical signal change. With a single membrane, the entire transmission range from 80 hertz to 33 kilohertz is covered. This means that in conventional multiple

Loudspeakers necessary crossovers and phase-overlapping overlaps in the sound-image-determining frequency range can be avoided.

Instead, the sound is created on a single, relatively small area, which in comparison to previous systems has come very close to the ideal of producing sound in one point. All the waves that define it form a complex sound pressure wave form in front of the membrane, as high and low frequencies occur simultaneously at different points of the membrane. Compared to conventional loudspeakers, the input signal is imaged almost perfectly so that the original and complete sound pressure image, which once reacted with the microphone, reaches the human ear.

Nevertheless, it has surprisingly been found that very sensitive listeners even when using bending wave transducers according to the principle just described, when they are located outside the stereo center, can not recognize clearly definable changes in the sound reproduction. Measurements have shown that this is directly attributable to the fact that the onset time to shifted even more and the rising edge of Pa to Pm - see Figure 2 - the more deformed, the farther the listener moves away from the central axis of the bending wave converter. This is shown in FIG. 4, in which the same measuring method as in FIG. 3 was used. The speaker was thus excited with a rectangular voltage and recorded its step response with a measuring microphone. In the arrangement of the measuring microphone in the central axis results in the course of the air pressure on Biegenwellenwandler over time, the curve A ₀ -B o- If the measuring microphone is then moved in increments of about 15 ° from the central axis, the measuring microphone are the Pressure curves A Bi up to A4-B4 measured. With increasing distance of the measuring microphone from the central axis to the extreme case of a distance of 60 °, the course of the pressure increase at the bending wave converter becomes more and more an S-curve. The increase is therefore not linear. FIG. 4 shows the measurement result on the time axis with a scale of 50 microseconds per raster unit. FIG. 4 shows that the non-linear course of the pressure rise of the curve A4-B ₄ extends over the period of approximately one grid length, that is to say of approximately 50 microseconds. In comparison with FIG. 1, it is clear that the "first time zone t0-ti" of human hearing with the

Sound source location after such a long time has almost passed. Therefore, even with a bending wave converter according to the principle of J.W. Manger perfect reproduction of the sound source location without another tool only possible lent when the listener is in the central axis of the bending wave transducer.

FIG. 5 shows the result of a same measurement as in FIG. 4, but here with an acoustic deflection element which covers one half of the bending wave transducer. Also with this

Measurement is the measuring microphone again first in the central axis the bending wave transducer arranged and then removed in each case in steps of about 15 ° from this central axis out. FIG. 5 shows quite clearly as a measurement result that the rise in the curve is absolutely linear independently of the respective angle.

This ensures that changes in the pressure value are reproduced undistorted, so that in this configuration of a Biegenwellenwandlers with one half set in front of acoustic deflector and sound movements from the first time zone of human hearing with the Schailquellenortung is correctly possible. In this configuration, the desired ideal of no more perceptible distorted reproduction of the sound motion in the time-accurate transformation of sound movements by mechanically moving air molecules in wire-bound electrical signals and back to sound movements is achieved by a designated as a loudspeaker transducer.

The presented here, comparatively very detailed description of the physical principles and the state of the art achieved so far has shown that conventional speaker systems and the previously known microphones can only transmit sounds and sounds. However, the object of this invention is a three-dimensional hearing in which also the vector information of each sound signal is transmitted, so that the listener can identify, in addition to the sounds and sounds, also their location and their nature,

This goal was almost fully achieved with the MSW bending wave transducers for converting electrical signals into the motions of air molecules. Only with the availability of such sound Converters can be shown the limits of known in the prior art microphones.

Widely known in numerous design variants are microphones which, as electroacoustic transducers, convert the motion impulses of the air surrounding them, which have been produced by sounds and sounds, into corresponding electrical voltage impulses.

It is indisputable that of all known variants, the principle of the so-called condenser microphone the least distortion in the

Conversion of the sound pressure image into an electrical voltage causes. In the current state of the art, a condenser microphone consists of a very thin, flexible and electrically conductive membrane, which is electrically insulated and mounted at a very short distance in front of a metal plate.

The membranes usually have only a material thickness of a few micrometers, so that they are still set in motion by very weak motion pulses of the ambient air. Between the membrane and the metal plate there is an air layer, which is compressed with each movement of the membrane. Thus, this air layer does not distort the effect on the movement of the membrane, the metal plate is usually provided with numerous holes through which the pressurized air flows and flows into a - also air-filled cavity behind the metal plate, compared to the air space between the membrane and Metal plate is very big.

Such an arrangement acts electrically like a plate capacitor with an air dielectric. In practice, capacities of about

20 to 100 pF typical. The motion impulses of the surrounding air also move the membrane, which changes the distance between the membrane and the metal plate and thus the capacitance of the capacitor.

In order to convert these variations in capacitance into an electrical voltage signal, the capacitor is connected via a high-resistance resistor to a DC voltage source, e.g. connected to a battery that charges it. When a motion pulse of the air is transmitted to the diaphragm and thereby moves, the capacitance of the charged capacitor changes and a current pulse flows from the DC voltage source over the capacitance and the resistance. A proportional voltage pulse can be tapped at the resistor and fed to a microphone amplifier. This microphone amplifier acts as an impedance converter and adjusts its impedance directly to the cable for signal transmission. The signal voltage is not amplified.

The resistance value must be set so high that with a change in capacitance in the rhythm of the lower limit frequency (for example 20 Hz) the charge is still sufficiently constant, so that the voltage on the capacitor changes with the sound vibrations. Depending on the capsule capacitance, resistance values of up to about 1 GΩ result.

Such a condenser microphone explains, for example, the published patent application DE 10 2008 013 395 A1. It describes how the metal plate opposite the diaphragm can be made of a material originally intended only as an electronic circuit board. On a mechanically very stable carrier material, eg made of ceramic, a thin copper layer is formed. brought, which can be brought into the required shape for a condenser microphone with little effort.

However, a single one of these microphones can by no means provide electrical signals for the task-oriented, three-dimensional hearing. It is well known that a separate signal is needed for each of the two human ears. Only by comparing the signals of all air movements formed in both organs of hearing can the human brain detect all the vector information in the sound.

Therefore, two microphones are required, which generate a total of two electrical signals. Each electrical signal is supplied to its own sound transducer, so a total of two sound transducers are driven. Such microphones are also known as stereo microphones.

A stereo microphone is the DE 91 01 371 before. It lists a total of three microphones, which are arranged side by side and can be brought into different angles to each other. If the microphones in a wink! aligned with each other, meet the scarf impulses only on a microphone frontally and on the other side. Since each microphone has a directional characteristic, the two signals differ significantly. This creates the disadvantage of mutual distortion.

If all microphones point in the same direction, the differences in the shape of the received signals are very small. A very serious disadvantage, however, arises with sound waves that do not impinge exactly orthogonally on the two microphones, but obliquely to their longitudinal axis. The microphones are at a distance from each other. of the. Even if the microphones touch mechanically, the distance in relation to the duration of the sound is still very large, within which evaluates the human ear a sound signal to its source, ie the transit time differences between the left ear and the right ear heard sound signal evaluated.

Only by detecting the transit time differences of otherwise left and right same signals, the human brain can derive the direction of the received signal - as previously explained in detail in this application as a so-called "first time zone of the perception of an acoustic pulse."

Therefore causes the delay difference according to the principle of DE 91 01 371 because of the distance of the microphones to each other - a dramatic for the detection of the direction of the sound signal distortion.

In the phase described in this application as a "second time zone of the perception of an acoustic pulse" in which the nature of the sound signal is evaluated, the human ear also evaluates level differences between the left and the right ear.

This phase of human "hearing" is dramatically falsified in the arrangement of two microphones according to the principle of DE 91 01 371, if the two microphones are not exactly alike and if they are not aligned exactly in the same direction Even an approach to this requirement very expensive and the result, however, never perfect. To mitigate this problem at least somewhat known on the current state of the art so-called coincidence microphones, in which two directional microphone capsules are arranged very close together in a common microphone housing. The two microphone capsules are rotatable relative to each other in their reference axis in the axis angle and usually have variable directional characteristics.

It is accepted as a principal drawback that even in the low and medium frequency range relative phase differences (phase differences) due to path differences at oblique

Sound incidence between two microphone channels occur. At high frequencies or with a very steep increase in the sound pulse, the phase difference is relatively much larger. Therefore, such coincidence microphones are usable only for pure, monocompatible intensity stereophony, which avoids extinction or impairment of frequency components in the electrical addition of the two stereo signals to a mono signal.

Against this background, the invention has the task of developing a stereo microphone that converts motion impulses and sound waves of the surrounding air into two electrical voltages with which - after appropriate amplification - a high-quality speakers - such as the bending wave converter according to the principle of JW Manger - Can be controlled and then the ambient air imparting motion impulses that not only produce music and sound perceived sound waves, but also generate air movements, with the help of which the human ear can locate the source of movement impulses in three-dimensional space and the nature of the sound signal at the beginning of the transient process can judge. The aim of the invention is therefore a stereo microphone, which converts pulses of motion of its ambient air into electrical voltages, which in turn another, electrical converter is controlled so that the air movements generated by it are identical to those air movements that activates the stereo microphone according to the invention during recording to have. It is therefore an essential part of the object of the invention to enable a three-dimensional spatial hearing.

The invention teaches that the Pvletallplatte is divided by a gap in a left metal plate as a left counter electrode and a right metal plate as a right counter electrode, which are electrically isolated from each other and each have their own electrical connection.

Thus, the mechanical form movements split by means of isomicrograph into two electrical form movements and thus into the desired dimensionally accurate and interference-free 3D stereo format.

With such a microphone according to the invention, it is possible to provide those electrical signals with which two high-quality loudspeakers can not only correctly reproduce music and sounds, but additionally provide the human ear with information that makes it possible to localize the source of the noise or the sound needed.

The 3D stereospalt microphone according to the invention thus closes the chain in the complete transmission, storage and subsequent reproduction of sounds and sounds. As explained in detail at the outset, the microphone according to the invention also absorbs the extremely rapid changes in shape that human hearing causes to erode. understanding of the vectorial nature of sounds and sounds.

The result is not just information to the listener, e.g. an orchestra on which side of the podium the timpani were set up and at what point on the stage the brass instruments stood, but also on the details of the sound information at the beginning of a sound and when switching from one sound to another. Only at the beginning of a sound pulse - namely in the first approximately 60 microseconds - does the brain's auditory center evaluate the information received with respect to the spatial orientation of the sound movement. Only in this period of time can the hearing detect the vectorial character of a sound image. The task of three-dimensional hearing becomes possible only with a microphone according to the invention.

It is essential that a microphone according to the invention receives and electrically reproduces this aperiodic, spatial information which is also at the beginning of each new periodic oscillation, that is continuously and repeatedly an essential part of the sound image. By detecting this "three-dimensional part", a microphone according to the invention enables complete recording and conversion of a complete sound image.

These ongoing small and very rapid changes in each tone - consisting of its root and its harmonics - determine what is often referred to as "timbre" or "character." It is only because of these features that it is possible to distinguish the sound of, for example, a Stradivarius from another ordinary violin. These properties of the 3D slit microphone according to the invention are based on the essential feature of the invention, namely that the metal plate of a conventional condenser microphone is divided by a gap in a left metal plate and a right metal plate. Both metal plates are electrically isolated from each other, so that from each metal plate own electrical signal for driving the respective speaker for the left and for the right side can be tapped. However, both metal plates have a single membrane in common. This results in two microphones, which have a previously considered impossible, tiny distance from each other and whose other differences are dramatically reduced compared to the prior art.

The very decisive and significant effect of the invention is that through the gap, the two metal plates and thus the two, directly juxtaposed microphones even the slightest differences in the duration of the sound can already be distinguished and recorded. Just as in the human ear, the two metal plates receive motion impulses that hit them directly on the front without any difference in transit time - ie exactly at the same time.

Because of the negligible differences between the two microphones in practice, the two signals generated by a 3D slit microphones according to the invention are exactly identical for exactly vertical sound pulses. With obliquely incident sound pulses, the differences in the transit time and in the level can be detected exactly. The source and species information is included in the first 30 milliseconds of each sound signal, whether it is a sound or a changing sound. In contrast to the prior art, this information is also contained in the first 30 milliseconds of the two electrical signals emitted by the 3D stereo microphone according to the invention. The smaller the gap between the two metal plates, the lower the propagation time differences for sound waves, which do not impinge exactly orthogonally on the surface of the two metal plates of the left and the right half of the diaphragm.

Therefore, it is advantageous that the width of the gap is as small as possible. However, in order that no electrical arcing occurs between the two metal plates over this small gap, the invention proposes, as a variant embodiment, that the gap between the two metal plates is completely filled by an electrical insulator. The material of this insulator should have the highest possible dielectric strength, but preferably should not increase the capacitance between the two edges of the metal plates, i. So do not have a high dielectric constant, which increases the total effective parasitic capacitance.

In practice, the capacity of each individual half of the microphone capsule is about 25-30 picofarads. On the other hand, the parasitic capacitance between the two front edges has an order of magnitude of 3-4 picofarads, ie about one-tenth of the capacity of the actual two microphone capsules.

This value causes such little "crosstalk" between the two channels of the SD stereo microphone according to the invention that in practice it does not matter can be achieved with the microphone according to the invention for the first time that the electrical signal contains not only the basic information about the motion pulses of the surrounding air, but also all those information when settling a sound that needs human hearing to the direction and thus the place as well be able to determine the type of sound signal quickly and accurately.

For this to be possible with the highest possible quality, a small width of the gap makes sense. In a first embodiment, the invention proposes a gap width of 100 microns. For even higher quality microphones according to the invention, the width of the gap is limited to a value between 40 microns and 10 microns.

A better, more helpful refinement is that the gap has the profile of a truncated cone, that is, is significantly narrower on the side of the two metal plates facing the membrane than on its opposite side.

A 3D stereo microphone according to the invention is constructed in the simplest embodiment as a pressure receiver, which has only one large opening, which is covered with a membrane. This achieves an approximately spherical directional diagram. Only for high and highest frequencies, the directional diagram has a preferred direction, so it is aligned with the sound source.

In one embodiment, a 3D stereo microphone according to the invention also has two openings in its hollow body, each of which is closed with its own membrane. Such a microphone capsule is called a "pressure gradient receiver" a set. This can be seen on the outside of the microphone because there are openings on both sides of the microphone capsule for the entry and exit of air. As a result, sound reflections which additionally act on the microphone, for example in interiors, are masked out, since only the difference between the pressure wave arriving from one side and the wave reflected from the other side is detected. Therefore, a Druckgradiente- nempänger basically a directional characteristic.

Of course, a microphone capsule according to the invention can also be used in all other known constructions of a condenser microphone to be presented.

In further embodiments, the microphone capsule of balls, cutting discs, ball plates, other plates, suspension devices for surfaces, sheets, moldings, spherical bodies, such as kidney, supercardioid or hypercardioid, separators, shotgun bodies, miniature bodies and / or other, known in the current state of the art surrounded by additional mechanical elements or physical forms. The appearance of a microphone according to the invention so its characteristic features are therefore not readily apparent directly, but mostly only after destruction of the microphone itself.

As a further embodiment, the two metal plates of the otherwise separate, two microphone capsules can be mounted on a common carrier. As a result, an increased mechanical stability and increased accuracy for the adjustment of the gap between the two metal plates are achieved. This also increases the accuracy of the proportion of the electrical signal, which after the conversion into sound movements of the air allows the human ear to localize, so makes up the three-dimensional portion of hearing. By a lifelike "in-form-make" thus increases the information content of the sound signal.

As a further improvement, the invention proposes that a partition wall adjoins the insulator in the gap and / or on the left metal plate and / or on the right metal plate near the gap and adjoins the microphone capsule. This achieves complete separation of the cavity behind the two metal plates. The invention recommends to separate the two chambers even air-tight from each other, because thereby any crosstalk from the "back" of the metal plates is avoided forth.

Then, each of these two chambers should be connected to the outside atmosphere via a small bore so that the air in the chambers does not stress the membrane like a spring, and thus air pressure changes do not squeeze or inflate the chambers like the can of a barometer.

In the most general case, the metal plates and the membrane may be oriented at any angle to each other. However, the invention prefers that the two metal plates are aligned parallel to the diaphragm, because then the errors in the conversion of the sound pressure into electrical voltages are lowest.

A preferred embodiment of the 3D slit microphone according to the invention has a membrane with an extremely low bending strength. The advantage is that the incident sound movements are distorted less and less with decreasing flexural strength of the membrane. Therefore, for the membrane in the xis a gold foil with a thickness of a few microns may be a preferred variant.

The shape of the hollow body serving as a microphone capsule is in principle arbitrary. However, for good transmission at all frequencies and at all slopes of a pulse, regular shapes are advantageous. The invention prefers that the microphone capsule is formed as a hollow cylinder whose lower end face is closed and whose upper end face is the opening which is closed by the membrane and the cross section of the hollow cylinder is a circle. However, since the metal plate is halved with respect to the membrane by a gap, it may also be useful to form the surfaces of both metal plates into an oval or an ellipse.

In any form of microphone capsule, it is advantageous if the gap passes through the geometric center of the microphone capsule. Thus, the gap is arranged in the region of the lowest bending strength of the membrane. As a result, the microphone according to the invention can also detect very small transit time differences between two sound movements.

The shape of the gap is arbitrary in the most general case. A grader gap reduces the parasitic capacitances between the two metal plates to the lowest possible level.

For the purpose of detecting even the smallest transit time differences between the two microphones of the 3D slit microphone according to the invention, the smallest possible gap is preferred which has the same dimensions over the entire length. The reduction of parasitic capacitance between the two adjacent metal plates However, this can be the reason for widening the gap to the outside and giving it the shape of a parabola, for example.

For the electrical connection of an SD gap microphone according to the invention, preferably between the electrical connection of the membrane and the respective electrical connection of the left metal plate and the right metal plate in series, an electrical energy source, e.g. a battery and a load resistor connected and tapped at each load resistor, an electrical signal that is supplied to an impedance converter and then a microphone amplifier.

A 3D slotted microphone according to the invention can also be produced as an electret microphone in a simpler embodiment. The advantage of an electret microphone over the condenser microphone is that it uses a permanent electrostatic polarization through an electret foil as a capacitor bias instead of an external supply voltage.

On the metal plates opposite the membrane an electret film is applied in each case, which provides for the membrane bias. The electret is fixed and the membrane is a metallised, lighter foil The size of the microphone capsule is usually between two millimeters and one centimeter The frequency response can in practice in a design as a pressure receiver - a microphone with omnidirectional characteristics - range from 20 Hz to 20 kHz Since, due to the electret, no high bias voltage is required for the diaphragm, in practice a voltage of about 1.5 V is sufficient for the supply of the impedance transformer. In a further, very interesting variant, a 3D stereospalt microphone according to the invention is integrated in a hearing aid. Here it could create dramatic relief, especially for the elderly, in which often an aging of the organ of equilibrium anyway generates orientation difficulties. For the function as a hearing aid, it is a very interesting embodiment that a inventive 3D split microphone z. B. is integrated in the front of a pair of glasses and thus allows a precisely directed hearing, so even the hearing impaired allows again to be acoustically oriented in the room.

Another very interesting application of a 3D slit microphone according to the invention is a vector microphone that absorbs sound velocity and sound pressure. This makes it possible to record the particle movement and its intensity.

In the following, further details and features of the invention will be explained in more detail by way of examples. However, these are not intended to limit the invention, but only to explain. It shows in a schematic representation:

FIG. 1: time profile of the air pressure at a measuring location when a dynamic pressure change is suddenly produced at a distance from it according to the current state of the art.

Fig. 2: different temporal pressure curves when using speakers with conventional electrodynamic transducers according to the current state of the art.

Fig. 3: a curve analogous to FIG. 2, but when using a loudspeaker with a bending wave transducer according to the current state of the art. Fig. 4: different waveforms analogous to FIG. 3 when using a loudspeaker with a bending wave transducer according to the prior art at different, not lying in the transducer axis measuring locations according to the current state of the art.

Fig. 5: Curves analogous to Figure 4, but in the half cover of the bending wave transducer with an acoustically impermeable deflection element according to the current state of the art.

6 shows a section through a 3D stereo microphone according to the invention

FIG. 6 shows the section through an SD stereo microphone according to the invention. The embodiment shown here consists of a hollow cylindrical microphone capsule 1, the underside of which is closed by a molded-on plane. The microphone capsule 1 has on its upper side the opening 11, which is surrounded in this embodiment by two circumferential projections.

On the lower projection in the opening 11, the metal plate 3 is mounted, which is the counter electrode to the diaphragm 2 as the first electrode. In Figure 6, the essential element of the invention is very clear and to recognize at first glance, namely the division of the metal plate 3 through the gap 4 in the metal plate left 3L and the metal plate right 3R. For the sake of clarity, the gap 4 in FIG. 6 is shown to be relatively large, although in practice it is so narrow, typically 10 to 40 micrometers wide, that it would only be identifiable as a dash if it were scaled in FIG ,

At the top of the opening 11, the second projection of the microphone capsule 1 runs around, on which the membrane 2 is fixed isolated. On average, it can be clearly seen that the membrane 2 runs parallel to the two metal plates 3L and 3R. It can also be seen very clearly that the diaphragm 2 as the first electrode and the metal plate 3L on the left as a counterelectrode and the metal plate 3R on the right as counterelectrodes each together with a part of the diaphragm 2 form a plate capacitor. The dielectric of these two plate capacitors, so the material between the first electrode and the respective counter electrode is in a 3D stereo microphone according to the invention air.

The function of a 3D stereospect microphone according to the invention is very easy to understand in FIG. If a sound movement provides for a movement of the air molecules and this movement propagates from air molecule to air molecule, then it will arrive after the end of the duration of the sound movement to a point of the membrane 2 and deform it. Depending on whether it arrives in the left region of the membrane 2 or in the right region of the membrane 2, it ensures a "dent" of the membrane 2. Thus, the capacity between the membrane 2 as the first electrode and the two metal plates 3L and 3R as counterelectrode.

FIG. 6 does not show that this change in capacitance is electrically interrogated by connecting a DC voltage source in series with a load resistor between the electrical terminal 2E of the diaphragm 2 and the electrical terminal 6L of the metal plate 3L and the electrical terminal 6R of the metal plate 3R , Any change in the capacitance between the membrane 2 and one of the two metal plates 3L or 3R can then be tapped off the load resistance as electrical voltage. In FIG. 6, for the sake of clarity, no additional material is shown in the gap 4 as an insulator.

Shown in Figure 6, the embodiment with a partition wall 5, which is connected in the embodiment shown here airtight with the microphone capsule 1, and also airtight with the respective edges of the two metal plates 3L and 3R. This creates two chambers hiring a single chamber as in known condenser microphones.

In Figure 6 it is shown that the two chambers are each connected by a small hole with the outside air to not be compressed or blown apart like a barometer can with changes in the atmospheric pressure of the ambient air. This bore also serves to escape air that would otherwise be compressed by a "bulging" of the membrane 2.

In FIG. 6, a fine perforation is shown as a variant of the metal plate on the left 3L and on the right 3R. Through these holes escapes the small air cushion between the membrane 2 and the two metal plates left 3L and right 3R.

In FIG. 6, the electrical connections 6L of the metal plate 3L and 6R of the metal plate 3R and 2E of the membrane 2 are shown in principle only. LIST OF REFERENCE NUMBERS

1 microphone capsule

11 opening

2 membranes

 2 E electrical connection of the membrane 2

3 metal plate

 3L metal plate left

 3R metal plate on the right

4 gap

 5 partition

 6L electrical metal plate 3L connection

6R electrical connection of metal plate 3R

Claims

claims

1. 3D stereospect microphone for converting sound movements of the ambient air into two electrical voltages consisting of

- a microphone capsule 1

 in the form of a hollow body,

 - Has an opening 11 which with a

 Membrane 2 is covered as an electrode,

 - Which consists of a flexible and electrically conductive film and

 - Which is fixed electrically isolated and which too

 - At least one metal plate 3 is spaced as a counter electrode, which is fixed within the microphone capsule 1,

 characterized in that

 the metal plate 3 is divided by a gap 4 into a left metal plate 3L as a left counter electrode and a right metal plate 3 as a right counter electrode,

 - which are electrically isolated from each other and

 - Each have their own electrical connection 61 and 6R

2. 3D stereospot microphone according to claim 1, characterized in that the hollow body has a further opening 1.

3. 3D stereospot microphone according to claim 2, characterized in that the hollow body has two, mutually opposite openings 11.

4. 3D stereospot microphone according to one of the preceding claims, characterized in that the microphone capsule 1 of

- balls and / or

 - Cutting discs and / or

 - Ball plates or other plates and / or

 - Suspension devices for interfaces and / or

 - Sheet and / or

 - Moldings and / or

 - spherical body and / or

 - Richtkörper such as kidney, supercardioid or hypercardioid and / or

- Separator and / or

 - Shotgun body and / or

 - Miniature body and / or

 other, known in the art of additional mechanical elements or physical forms is surrounded.

5. The 3D stereo microphone according to one of the preceding claims, characterized in that between the left metal plate 3L and the right metal plate 3R, an electrical insulator is installed, which can completely fill the gap 4 between the two metal plates 3L and 3R.

6. 3D stereospot microphone according to one of the preceding claims, characterized in that the width of the gap 4 is smaller than 100 microns.

7. 3D stereo gap microphone according to one of the preceding claims, characterized in that the width of the gap 4 is smaller than 40 microns and larger than 10 microns.

8. 3D stereo gap microphone according to one of the preceding claims, characterized in that the gap 4 between the two metal plates 3L and 3R near the diaphragm 2 is narrower than on the opposite side of the metal plate 3L and 3R.

9. 3D stereospot microphone according to one of the preceding claims, characterized in that both metal plates 3L and 3R are mounted on a common carrier

10.3D stereo gap microphone according to one of the preceding claims, characterized in that a partition wall 5 adjacent to the insulator in the gap 4 and / or to the left metal plate 3L and / or to the right metal plate 3R near the gap 4 and to the microphone capsule. 1 followed.

11.3D stereo gap microphone according to one of the preceding claims, characterized in that the partition wall 5 is connected airtight at its edges, so that it divides the microphone capsule 1 into two pneumatically separated halves.

 1L and 1R

12.3D stereo gap microphone according to one of the preceding claims, characterized in that the metal plates 3R and 3L are aligned parallel to the membrane 2.

13.3D stereo gap microphone according to one of the preceding claims, characterized in that the membrane 2 has the lowest possible bending strength

14.3D stereo gap microphone according to one of the preceding claims, characterized in that the microphone capsule 1 is formed as a hollow cylinder whose lower end face is closed and whose upper end face is the opening 11 which is closed by the membrane 2 and the cross section of the hollow cylinder

 - a circle or

 - an oval or

 - is an ellipse.

15.3D stereo gap microphone according to one of the preceding claims, characterized in that the gap 4 extends through the geometric center of the microphone capsule.

16.3D stereo gap microphone according to one of the preceding claims, characterized in that between an electrical connection 2E to the membrane 2 and the two electrical connections 6L and 6R of the left metal plate 3L and the right metal plate 3R in series depending on an electrical energy source such as a battery and a load resistor are connected and at each load resistor, an electrical signal can be tapped, which can be supplied to an impedance converter and then a microphone amplifier.

17.3D stereo gap microphone according to one of the preceding claims, characterized in that the gap 4 between the two metal plates 3L and 3R is straight and over the entire length can have the same profile.

18.3D stereo gap microphone according to any one of the preceding claims, characterized in that the gap between the two metal plates 3L and 3R increases towards the outside and has for example a parabolically shaped edge.

19. 3D stereospalt microphone according to one of the preceding claims, characterized in that on the membrane 2 opposite metal plates 3L and 3R, in each case an electret foil is applied, which causes a membrane bias.

20.3D stereo gap microphone according to claim 19, characterized in that the electret foil is spatially stationary and / or the membrane is a metallized film.

21 .3D stereo gap microphone according to one of the preceding claims, characterized in that it is integrated in a hearing aid.

22. Application of a 3D stereospalt microphone according to the invention as a vector microphone, characterized in that it receives sound velocity and sound pressure.