EP2139265A1 - Motor for a tweeter - Google Patents

Abstract

The motor has magnets (6) transferred towards a rear or behind the motor by pivotment, and aligned with a coil (3) of transducer. Pole pieces e.g. field plate (9) and core (7) and magnets follow the pivotment. The plate, magnet and core assembly are coaxial around a vertical axis of symmetry. The assembly is in the shape of tube or truncated cone, and has tailpipe and venturi effects by addition of metal e.g. soft iron, with the core, in a radial magnetization direction to release air space (ZC) under an inverse dome shaped diaphragm (1) and authorize spaces of magnet and metal in the pieces.

Description

Technical sector of the invention:

The present invention relates to the technical field of loudspeakers in particular high and very high range and in particular their component type "tweeter" ("acute" transducer, as is known).

Technical problem posed:

In the field of loudspeakers, particularly high or very high range, it is essential to achieve very fine and very difficult compromises between the characteristics of weight, rigidity, damping of acoustic waves at the membrane (depending on various parameters of which naturally the speed of sound propagation).

In general, and in this context, the "acute" or " tweeter " transducer is a very important element in the design of any acoustic reproduction system.

• Il couvre la bande de fréquence la plus large possible (exemple non limitatif : les 2,5 kHz à 40 kHz du « tweeter » béryllium (Be) de la gamme d'enceintes acoustiques très haut de gamme Utopia ™ Be de FOCAL JMIab ™).
• Il couvre la bande de fréquence dans laquelle on trouve les dernières harmoniques (11) des « sons fondamentaux » (272 à 3636Hz), avec en particulier la reproduction des harmoniques de rang 2 et 3 (1 kHz à 20 kHz).

For the record, the role of acute in an acoustic reproduction system is characterized by the following main points:

• It covers the widest possible frequency band (non-limiting example: the 2.5 kHz to 40 kHz of the beryllium tweeter (Be) of the FOCAL JMIab ™ Utopia ™ Be high-end loudspeaker range) .
• It covers the frequency band in which we find the last harmonics (11) of the "fundamental sounds" (272 to 3636Hz), with in particular the reproduction of harmonics of rank 2 and 3 (1 kHz to 20 kHz).

This last point is crucial: indeed, the human ear being a "transient type" sensor, it is the perception of all the harmonics that will give the restored sound all its finesse, its definition.

As can be seen, the reproduction quality of the "acute" transducer is an important factor in the passage of micro-information, useful for the subtleties of listening in stereophonic (or higher).

In 2002 - 2003, FOCAL JMlab ™ created the Grande Utopia ™ Be benchmark that will have a worldwide impact. For the first time, pure beryllium dome broadband tweeter technology was implemented. The trade press quickly seized the event and the Grande Utopia ™ Be immediately became the best speaker in the world for them, highlighting the extraordinary performance of an extraordinary "tweeter".

The totally innovative use of beryllium has set a new standard in performance and musicality: low distortion rate, linear response, extended response to more than 40 kHz, ...

The Applicant is the first and only designer / builder to have introduced Beryllium in its purest form and this despite a strong prejudice that was unfamiliar with all the technical difficulties associated with its industrialization.

In order to go further in the source point, the Applicant has decided to engage later an inventive step focused this time on the acoustic connection between the HP ("speaker") of medium and the "tweeter" . The plaintiff started from the idea that the localization of sound by the ear is done mainly by high order harmonics which are therefore reproduced by the "tweeter" (important notion of "source point").

The objective of the present invention is the creation of a broadband "tweeter" that is to say covering 4 octaves, high efficiency between 95 dB and 98 dB, allowing full power operation from 1 kHz to 40 kHz, preferably from 2 to 35 kHz, with very good power handling, low distortion, low directivity and despite these advantages and modifications is able to remain consistent with the requirements of very high quality .

The Applicant has sought to produce a high-end tweeter transducer which, as the Applicant discovered, must have a high efficiency to increase dynamic capacity and power handling, and the most important inventive merit has been thinking of designing a completely original engine structure.

• avoir une fréquence de résonance basse pour réduire les colorations sonores et la distorsion ;
• avoir un taux de distorsion inférieur à 0,5% à 2000 Hz ;
• maintenir le faible taux de distorsion aux fréquences supérieures à 2500 Hz (taux actuel à 0,25%) ;
• avoir une très bonne courbe de réponse linaire en fréquence de 1 kHz à 40 kHz, avec un amortissement rapide et une absence de résonance ;
• avoir un rendement supérieur ;
• posséder une bonne réponse transitoire ;
• une très bonne richesse harmonique ;
• une faible « coloration »;
• l'ensemble doit être compact pour des questions de facilité d'intégration de « tweeter » dans l'enceinte ;
• les coûts de production doivent rester maîtrisés.

The criteria are:

• have a low resonance frequency to reduce sound staining and distortion;
• have a distortion rate of less than 0.5% at 2000 Hz;
• maintain the low distortion rate at frequencies above 2500 Hz (current rate at 0.25%);
• have a very good linear frequency response curve from 1 kHz to 40 kHz, with fast damping and no resonance;
• have a superior return;
• have a good transient response;
• a very good harmonic richness;
• a weak "color"
• the set must be compact for questions of ease of integration of "tweeter" in the enclosure;
• production costs must be kept under control.

One objective was to improve the performance of the "tweeter" and its performance to improve the power handling, improve the acoustic quality, the level in the extreme acute.

Those skilled in the art will immediately understand the complexity of the problem, and the notoriously antagonistic nature of the improvements sought.

The concepts of transducer, midrange, treble, bass, harmonic, rank of harmonics, power, power handling, performance, resonance, distortion, dragging, membrane, coloration (modifying the exact sound of a given instrument by giving it , when passing through the transducer, an "acoustic coloration" very slightly different, that the music lover detects), bandwidth, acoustic connection, transparency (ability to make the sound with perfection and no detectable effect of veil), point source, "waterfall (Response curve (in symbolic form of "waterfall") in frequency in time: the extinction of a given frequency - and of all the frequencies of the band - must be as fast as possible, no "Dragging", and must be ideally instantaneous, that is to say without "extinction accidents" and thus without bringing "color" to a given instrument; Figure 15 etc., etc., as well as HP, "tweeter" and the like are well known to those skilled in the art and will not otherwise be defined here. This is general knowledge of the skilled person.

The Applicant has managed to design a completely original engine that makes it possible to manufacture a "tweeter" with a very wide band, preferably a Beryllium membrane, which operates in almost infinite load; therefore, its resonant frequency is very low since it is 700 Hz. This "tweeter" can operate from 2000 Hz to 40000 Hz with a low rate of distortions. The tests were carried out on a FOCAL ™ Electra ™ 1000 enclosure.

It was noted that the beryllium membrane (Be) certainly provides a piston operation of the membrane but also that the "tweeter" has nevertheless much coloring and distortion below 2000 Hz to be used at this frequency. This is related to the presence of a mechanical resonance frequency (spring mass) too close to the bandwidth used. This first point shows the great difficulty of the problems and the great sensitivity of the various parameters.

In this context, the ability to cut the "tweeter" lower in frequency is a considerable contribution in terms of phasing with other transducers, in terms of micro detail, stereophonic image, and transient compliance. These notions are also very well known.

The objective value of minima 2 kHz as a low frequency point which can be used in practice is a very sensitive point. It impacts the connection between the "tweeter" and the "medium" transducer. If one tries to get down to 2 kHz, there is a problem of power handling of the "tweeter" and against it avoids the medium transducer to become too directive. Indeed, it is necessary to manage the dispersion and acoustic radiation at the connection frequency, that is to say obtain a similar dispersion diagram for the two transducers so that the effect is not noticeable.

Unfortunately, another disadvantage has arisen, which is that the thus obtained enlargement of this bandwidth also results in an increase in the effective power absorbed by the acute transducer. This may reduce the power of the "tweeter" but also compress the level on high dynamic signals and is therefore not acceptable.

The Applicant then considered increasing the efficiency of the transducer to maintain all the qualities.

It has been possible to progress in terms of performance with the development of a rigid and ultra light 60 mg membrane. Today, it seems difficult to reduce the weight again on pain of having a membrane too fragile and too difficult to manufacture. However, this improvement of the membrane certainly provides a very significant extra, but is a preferred option and therefore not mandatory. The capital point of the invention, as will be seen below, is the engine design.

It was then judged that it was necessary to try to increase the yield by increasing the force factor, and this was only possible after many unsuccessful attempts by creating a magnetic motor of entirely original design, more powerful and compatible with a strong decompression of the membrane.

The Applicant has deliberately avoided the use of mechanical resonances which, however, artificially increase the sound level (higher efficiency but artificially, especially using a high resonance frequency and "break up" at 20 kHz not amortized) but which generate a lot of dragging.

As an example of difficulties encountered, the response curve was initially, during the tests, improved with the removal of the mechanical overvoltage at the beginning of the band and a response curve with few accidents. The damping of the beginning of the tape was excellent, but at the expense of the level of distortion.

The Applicant has therefore continued its research to solve (among other problems) this problem of distortion and further improve the performance of the "tweeter" by increasing the performance.

Prior art:

Those skilled in the art will find below the details of the state of the art.

Transducers using a compression chamber loaded by a horn (adaptation of the acoustic impedance).

Remember that a flag is an adapter for concentrating sound energy in one direction (a very simple analogy would be the megaphone). This type of solution is very harmful in "high fidelity" where precisely it is not necessary to have a "directed" energy and where on the contrary it is necessary to produce a "three-dimensional" effect (whereas a flag is obviously not three-dimensional since it is "Directive" in sound energy).

Such a system is known to have the advantage of having a lot of efficiency (110 dB / 1watt / 1m) but this load is very cumbersome (about 30 cm in diameter and almost the same in depth).

The directivity is also very pronounced and the bandwidth in the high spectrum is reduced to 19kHz for the best. Which therefore reserves the sound of high power. This system is not suitable for high or very high fidelity high fidelity for which it is essential not to be directive and where it On the contrary, it is necessary to reproduce or recreate the spatial dimension of the stereophonic message.

Some manufacturers use electrodynamic "tweeters" with a high resonance frequency. This makes it possible to very tacitly increase the sound level at the beginning of the band from 4 to 6 dB. To increase the high of the treble it combines a frequency of "break up" (deformation and possibly rupture) of the membrane below 20 kHz and not damped. These mechanical resonances enormously increase the drag and reduce the bandwidth.

Other manufacturers add to this flag primers.

Some manufacturers use a lower impedance (3 ohms instead of 6 ohms). This changes the sensitivity but the performance remains the same. The coil uses more current and creates problems of overheating, sound compression, distortion of the amplifier, etc ...

These techniques must be considered as artifices because the actual efficiency depends solely on the emitting surface, mass of the moving equipment and the force factor of the coil.

Here again we see that the yield is an antagonistic property of others also sought, hence again the difficulty of finding a satisfactory compromise, and even more so a solution which, as in the invention, reduces to a maximum compromises "harmful.

• des « équipages mobiles » (essentiellement l'ensemble en vibration constitué par la bobine située dans l'entrefer et la membrane solidaire de la bobine, et éventuellement la « suspension » de la membrane, c'est-à-dire la partie périphérique de raccordement de la membrane au support de membrane) a priori plus légers et plus rigides ;
• des aimants plus puissants pour le moteur.

Overall, the achievement of "tweeter" having a high efficiency is, in the prior art, necessarily very expensive since the materials used are more efficient:

• "mobile crews" (essentially the vibration assembly constituted by the coil located in the gap and the membrane integral with the coil, and optionally the "suspension" of the membrane, that is to say the peripheral portion of connection of the membrane to the membrane support) a priori lighter and more rigid;
• stronger magnets for the motor.

Most manufacturers therefore use the following tradeoff: just enough yield to get a minimum of quality and reduce their production costs.

On the contrary, the Applicant directed his research towards high-performance acute transducers. It is therefore a totally different technical approach, and infinitely more difficult.

• En 1991 le Tn 46™ avait un rendement de 85 dB/1watt/1 m.
• En 2005 le TBU™ avait un rendement de 94 dB/1watt/1m.

As an example of the Applicant's prior art:

• In 1991 the Tn 46 ™ had a yield of 85 dB / 1watt / 1 m.
• In 2005 the TBU ™ had a yield of 94 dB / 1watt / 1m.

• aimant Ferrite de plus en plus gros diamètre 60 -> diamètre 96 ;
• arrivée des aimants hautes puissances : ferrite -> néodyme ;
• utilisation du Télar 57™, Focus ring™ (réduction d'une partie de fuite magnétique).

The Applicant has innovated in the following sectors:

• ferrite magnet larger and larger diameter 60 -> diameter 96;
• arrival of high power magnets: ferrite ->neodymium;
• use of the Telar 57 ™, Focus ring ™ (reduction of a magnetic leakage part).

An additional constraint concerns the rear clearance of the membrane, that is to say the confined volume "under" the membrane and "above" the magnets, where the air must be able to circulate freely enough and escape in order to not to create compression that is to say a parasitic force acting deleteriously on the membrane. Some "tweeters" called "decompressed" appeared.

These examples not only show the Applicant's powerful, ongoing research effort, but also make it clear that progress is difficult and innovations are increasingly problematic because they are cutting-edge technologies. In such areas, often, even a "weak" improvement of a property can represent a significant inventive effort as well as the struggle against many technical or "historical" prejudices, and the difficulty of always allowing contradictory or incompatible factors. to work, finally, within the same innovation.

Technical difficulties encountered.

The characteristic curves of the ferrocobalt metal for the pole pieces (field plate and core) in the FEMM software or simulation software did not exceed 130,000 (mOe) when it was used until saturation. It was not possible to measure above 130,000 (mOe) because it requires very powerful amplifiers. The inventors were therefore compelled to calculate the values above, up to 240 000 mOe for the program to go to the level of real saturation.

Difficulty finding a compromise between decompression level and engine performance.

It could have increased the performance of the engine but the inventors would have been forced to reduce the decompression of the engine. This would have degraded the response curve of the "tweeter".

This is another example of the difficult tradeoffs mentioned above as one of the major obstacles in this technology.

Various difficulties set out above with regard to the realization of a very superior compromise, with a "quantum leap", to the very average compromises of the prior art.

• difficulté d'assemblage des pièces du moteur notamment en coaxialité et du fait des petites dimensions des pièces,
• nécessité (cf. ci-dessous) d'aimanter l'aimant dans sa nouvelle configuration en « tube » ou « cône » (un tube ou cône est impossible à être aimanté en une pièce).

Essentially two very important difficulties:

• difficulty in assembling the engine parts, especially in coaxiality and because of the small dimensions of the parts,
• need (see below) to magnetize the magnet in its new configuration in "tube" or "cone" (a tube or cone is impossible to be magnetized in one piece).

Detailed description of the invention

According to its general concept, the invention proposes a completely original engine, and its incorporation in particular a "tweeter" which provides all the benefits referred to in the technical problem specified above, while still achieving some compromises still inevitable to date but in proposing a solution that achieves a "quantum leap" with respect to the prior art.

• la membrane est supposée définir globalement l'horizontale par la ligne ou le plan reliant les bords opposés de la suspension
• elle est supposée par convention être placée vers le haut, moteur étant donc sous la membrane
• la membrane étant horizontale, elle définit donc également une direction verticale qui est perpendiculaire à son plan

The spatial references taken are the following:

• the membrane is supposed to globally define the horizontal by the line or the plane connecting the opposite edges of the suspension
• it is supposed by convention to be placed upwards, so the engine is under the membrane
• the membrane being horizontal, it therefore also defines a vertical direction which is perpendicular to its plane

As will be seen below, the present invention covers in particular a motor for a loudspeaker and especially for a "tweeter"("acute" transducer), characterized in that the magnet 6 of the motor is transferred "towards the rear" or "behind the engine" and is adapted as well as the pole pieces (field plate 9 and core 7) to release the volume under the membrane and remove or almost any compression while allowing either a larger volume of magnet or a volume more metal in the polar pieces, or both.

• l'aimant, au lieu d'être disposé sensiblement « à plat » ou horizontalement sous la membrane, c'est-à-dire positionné à côté de la bobine, cf. aimant 6 figure 1 A art antérieur, est aligné avec la bobine du transducteur et avec les pièces polaires 9 (plaque de champ) et noyau 7, juste au-dessous de la membrane, l'ensemble adoptant donc une configuration en tube ou en cône tronqué « verticale » et « sous » la membrane, tout en conservant naturellement les impératifs géométriques d'un moteur, comme notamment la position relative des pièces, de l'entrefer, etc...
• est, selon l'invention,
• disposé manière verticale ou sensiblement verticale sous la membrane, ce qui forme un tube ou un cône tronqué, avec de plus la pièce polaire 9 déportée à l'extérieur, cf. figure 1B ou 1C ou 1D ou figure 2, avec enfin le noyau 7 déporté vers l'intérieur.

By "behind the engine" is meant the fact that:

• the magnet, instead of being disposed substantially "flat" or horizontally under the membrane, that is to say positioned next to the coil, cf. magnet 6 figure 1 A prior art, is aligned with the transducer coil and with the pole pieces 9 (field plate) and core 7, just below the membrane, the assembly thus adopting a "vertical" truncated tube or cone configuration and "Under" the membrane, while naturally maintaining the geometrical requirements of an engine, such as the relative position of the parts, the gap, etc ...
• is, according to the invention,
• disposed vertically or substantially vertically under the membrane, which forms a tube or a truncated cone, with the pole piece 9 further away, cf. Figure 1B or 1C or 1D or figure 2 , with finally the core 7 deported inward.

Thus, a "pivoting" of the magnet has been performed, and pole pieces 7 (or at least part of the core 7) and the field plate 9 under the membrane (or more generally under the moving element). ) (these elements must follow the "pivoting of the magnet to continue to produce an effective magnetic field) which forms a totally concentric or coaxial" vertical "motor, and releases volume at the compression zone ZC and especially releases around it any obstacle. The magnetic field is also pivoted, the direction of magnetization becoming radial.

It will also be seen that the pole pieces (core 7 and field plate 9) are correspondingly adapted to also form a truncated tube or cone. This will be understood by those skilled in the art since, after "pivoting the magnet," it is still necessary to generate an effective magnetic field for driving the coil ("radial" magnetization).

The assembly thus forms a cylindrical tube, or a truncated cone, possibly with a nozzle effect produced by the pole piece 7 (core) whose inner shape may optionally adopt a venturi profile, cf. figure 1D , Fig. 2 . Said inner profile is dictated solely by the amount of metal to be placed in the pole piece 7, this amount being dictated by the distance from the gap. In fact the saturation increases with the proximity of the air gap, and it is therefore advantageous to place more metal when one is near the air gap: an elegant solution is the domed shape or also called "venturi"; a good magnetic optimization is obtained by reducing leaks or magnetic losses while maintaining a more transparent motor.

Indeed a conical shape reduces the magnetic leakage or loss, but the cylinder shape is preferred because it also removes a large part of magnetic leakage while avoiding the risk of agreement at certain frequencies.

It can still be noted that this arrangement favors two options that each have advantages, the first being to have more magnet, and the second is to have more metal in the volume released by the new provision. We can also add a little magnet and a little metal, that is to say, combine the two options.

The decompression vent 8 becomes completely optional according to the invention since the air of the zone ZC can be evacuated freely by the periphery, cf. arrows on the Figure 1C .

Such a motor characterized in that said magnet is divided into several segments of magnets 6A, 6B, 6C, ... arranged vertically or at an angle relative to the vertical (longitudinal axis of the engine), in particular magnets in position vertically collinear around the core 7, and the field plate 9, leaving an air gap 2 to accommodate the coil 3.

Such a motor characterized in that said segments form a truncated tube or cone possibly with a "nozzle" or "venturi" effect around the core 7, as can be seen in the accompanying figures, the nozzle effect being explained below. on top to bring more metal to the core and thus move and move the place away from saturation.

Such a motor characterized in that the polar core and field plate parts and the magnet segments form a "vertical" tube or truncated cone, under the membrane taken as a horizontal reference, cf. anxious figures.

Such a motor characterized in that the nature of the magnet is neodymium, in particular N38M (Br 12.2 kGs) or optionally N48M (Br 13.6 kGs).

Such an engine characterized in that the metal of the pole pieces can be ferrocobalt.

Such an engine characterized in that the metal of the pole pieces is a soft iron E24.

And naturally the technical equivalents of such engines, using materials and equivalent forms directly accessible to the skilled person.

The invention will be better understood on reading the description which follows, and non-limiting examples below.

Trials necessary for the development and resolution of difficulties and / or the achievement of viable "technical compromises".

It was necessary to establish an optimization of the mechanical parts and with a good compromise and we obtained a decrease of the distortion at 2 kHz to 1%, that is to say very important.

The thickness of the thinned surface by the decompression remains the same: 1,5m.

The area thinned by decompression was reduced to a diameter of 64 mm.

The dimensions of the engine mounting brackets have been increased to 8x12 mm.

Other optimizations are possible, playing differently on the various factors above as will be understood by the skilled person.

• une réduction de la surface de décompression de 437 mm²;
• une baisse de la distorsion à 2 kHz ; elle est à présent à 1 % ;
• une légère dégradation de la courbe de réponse avec une ondulation de +/-1 dB, visible également sur la courbe d'impédance.

The results are as follows:

• a reduction in the decompression area of 437 mm ² ;
• a decrease in distortion at 2 kHz; it is now 1%;
• a slight degradation of the response curve with a ripple of +/- 1 dB, also visible on the impedance curve.

This result shows that, in this technology, a great difficulty is that fine pieces and large sizes can generate significant distortions in the treble. This is all the more delicate when it is necessary to maintain a large decompression.

After the modifications and compromises above; comparative tapping of the "tweeter" decompressed or reinforced with the uncompressed "tweeter" has been carried out. They confirmed the advantage of decompressing the "tweeter" compared to a "tweeter" closed yet offers more transparency in the bottom of the "tweeter". The suppression of mechanical resonances makes the bottom of the treble more silky.

One of the crucial objectives was to improve the performance of the "tweeter" and its performance, to improve the power handling, improve the acoustic quality, the level in the extreme acute.

• 7.7 ohms à 20 kHz contre 6,5 ohms à 4 kHz.

During the research and testing, it was noticed on the impedance curve of the "tweeter" that the electrical resistance at 20 kHz is higher than the rest of the band.

• 7.7 ohms at 20 kHz against 6.5 ohms at 4 kHz.

This increase of the electrical resistance makes that the "tweeter" consumes less current at this frequency which leads, another problem, to a loss of sensitivity.

It has been sought to reduce this increase in electrical resistance by reducing the inductance of the coil, in order to gain level in the high frequencies.

By way of illustration and not limitation, but representative, the coil thus developed has an inductance of 0.03 mH, an electrical resistance of 6 ohms and a wire diameter of 0.08 mm.

Calculations and research had to be done to reduce the inductance of the coil and this resulted in a reduction in the conductor wire diameter.

This reduction in diameter reduces the mass of the coil but also reduces the magnetic induction

Unfortunately the loss of mass does not compensate for the loss of magnetic induction because the mass of the membrane and the mass of the suspension are still present.

Overall this does not translate into an overall loss of level.

This is another example of difficulty in applying "logical" solutions.

In addition, this reduction in the diameter of the wire may reduce the power handling of the "tweeter" by reducing the copper mass and the fragility of the copper wire (confirmed by the coil manufacturer).

The inventors, faced with this failure encountered during the implementation of a possible solution, had to try another solution to reduce this increase in electrical resistance at high frequencies.

The use of a Faraday ring has made it possible, among other things, to reduce the inductance of the coil at high frequencies, by the self-induction of an electric current in the Faraday ring that generates a magnetic field that modifies the magnetic field. inductance of the coil.

Further tests were then performed by adding a second 0.05 mm thick Faraday copper ring to the core and the field plate.

The air gap of the engine has not been modified to not lose a magnetic field.

There was no change in the level of the "tweeter", a new failure of a logical solution. The level of the "tweeter" is identical to the un-banded motor on the entire band of the "tweeter".

The second ring seems to cancel the effect and it has been renounced.

After removing the Faraday ring from the field plate (the motor air gap was not modified to avoid losing magnetic field), only a small level increase in the frequencies of 0.4 dB from 15 kHz up to 40 kHz.

The inventors thought that the thickness of the ring was too thin to allow a good efficiency of the Faraday ring (intensity of the current in the ring). An increase to a value of the order of 0.2 mm appeared necessary.

To insert a 0.2 mm thick Faraday ring, it was necessary to modify the motor core to compensate for the over-thickness of the ring. As a result, the gap has been increased.

Increasing the thickness to 0.2 mm obviously increased the effect of the Faraday ring.

The impedance measurement then made it possible to note a reduction of the resistance at 20 kHz, since it went from 7.7 ohms to 7.2 ohms. The effect is also visible on the response curve with an increase of 1 dB from 15 kHz to 30 kHz, and from 2 dB to 40 kHz.

But the inventors have noted that despite their expectations, the effect of the ring can not compensate for the loss of magnetic field created by the increase in the air gap of the motor, a loss of the level of the "tweeter" of 1 , 5 d8 on the whole band.

New failure of a logical solution.

The inventors then sought a radically different solution.

Increase the magnetic power of the engine while maintaining the important decompression of the engine

One difficulty was to increase the power of the engine without increasing the diameter of the engine, to maintain the strong decompression necessary for the proper functioning of the "tweeter".

Considering it impossible to increase the performance without increasing the diameter of the motor, a radical solution was tried and contrary to the teaching of all the prior art, that is to say to transfer the magnet behind the motor. This requires changing the orientation of the magnet.

This however generates free volume to add a larger magnet volume but the behavior of the magnetic circuit in this position obviously becomes unpredictable. One of the merits of the Applicant is to have nevertheless continued in this already original way in itself.

• l'aimant utilisé est du néodyme : N38M ;
• le métal utilisé est un fer doux : E24 ;
• longueur de la plus petite boucle du circuit : 14,7 ;
• surface d'échange entre l'aimant et le E24: 7,8 cm3 ;
• quantité d'aimant : 6,1 cm3 ;

Prototype:

• the magnet used is neodymium: N38M;
• the metal used is a soft iron: E24;
• length of the smallest circuit loop: 14.7;
• exchange surface between the magnet and the E24: 7.8 cm3;
• amount of magnet: 6.1 cm3;

The skilled person will consider other combinations around this one by following the teachings of this description and his own general knowledge.

Series of magnetic simulations (with a soft iron E24) Simulation of the transfer of the magnet behind the motor changes the magnetic orientation.

This amounts to moving from a ring-shaped magnet with the direction of axial magnetization to a tube-shaped magnet with a radial magnetization direction .

See Figure 1 attached.

• 1A art antérieur schématisé à aimants positionnés à côté de la bobine du transducteur c'est-à-dire derrière la membrane. Le sens d'aimantation est alors axial.
• 1B moteur de l'invention schématisé à aimants en position verticalement colinéaires, vers la partie arrière du moteur, c'est-à-dire derrière la bobine du transducteur. Le sens d'aimantation est alors radial.
• 1C coupe AA de la figure 1B, moteur de l'invention montrant schématiquement d'autres composants, dont l'équipage mobile.
• 1D moteur de l'invention, en coupe transversale (prototype).

The figure 1 consists of the figures:

• 1A prior art schematized magnets positioned next to the coil of the transducer that is to say behind the membrane. The magnetization direction is then axial.
• 1B engine of the invention schematized magnets in position vertically collinear towards the rear part of the engine, that is to say behind the coil of the transducer. The magnetization direction is then radial.
• 1C AA cut of the Figure 1B , engine of the invention schematically showing other components, including the moving equipment.
• 1D motor of the invention, in cross section (prototype).

1

membrane (ici mode préférentiel en « dôme inversé »)

2

entrefer

3

bobine

4

suspension de la membrane

5

support de la membrane (et de l'équipage mobile)

6

aimants

7

noyau

ZC

volume d'air sous la membrane (où se produit une compression dans l'art antérieur)

8

art antérieur : orifice de décompression rendu nécessaire par la compression sous la membrane dans la zone de compression ZC.

9

plaque de champ

In all the figs, throughout the description and the claims, the same references have the same meanings, unless otherwise stated:

1

membrane (here preferred mode in "inverted dome")

2

gap

3

coil

4

suspension of the membrane

5

membrane support (and mobile equipment)

6

magnets

7

core

ZC

air volume under the membrane (where compression occurs in the prior art)

8

prior art: decompression port made necessary by the compression under the membrane in the compression zone ZC.

9

field plate

It is recalled that the moving element is formed of the coil 3 which oscillates in the "suspended" position in the gap, under the action of the electromagnetic forces, and by the membrane 1 which vibrates under the action of the coil to which the membrane is attached mechanically.

It is noted that in the prior art zone ZC is an airtight volume where inevitably occurs a variable compression very detrimental to performance. The prior art often overcomes this problem by a decompression orifice 8 whose effect is weak.

With such a configuration, it is not possible, in the prior art, to go down in frequency .

The magnet used for the prototype of the invention, Figure 1D , is neodymium: N38M
The metal used for the pole pieces is a soft iron: E24
Length of the smallest circuit loop (magnetic field): 10.82 mm Magnet quantity: 6.05 cm3.

Simulations with the known GEMINI ™ simulation software show that the invention surprisingly achieves the same magnetic power with almost the same magnet amount of 1.85 tesla.

This is very interesting because it opens up the possibility of increasing engine performance knowing that we still have room available behind the engine.

Optimization of the magnetic circuit to have a better compromise between:

This example shows that the parameters can be optimized differently, and the person skilled in the art thus has, and with the following examples, markers for making his own prototype.

• Les performances magnétiques (quantité d'aimant, longueur de la plus petit boucle du circuit)
• Décompression (quantité de matière)

The Claimant played on:

• Magnetic performances (quantity of magnet, length of the smallest loop of the circuit)
• Decompression (amount of material)

See Figure 2 attached

• The magnet used is neodymium: N38M;
• The metal used is a soft iron: E24;
• Length of the smallest loop of the circuit: 7.75 mm;
• Amount of magnet: 10.5 cm3;
• Exchange surface between the magnet and the E24: 25.3 cm3;
• Decompression area is 1,115 mm ² .

The results obtained are:

See appended FIG. 3, which consists of FIGS. 3A to 3C.

• average field in the air gap: 2.1 teslas;
• maxi fields in the metal: 2,28 teslas;
• engine price: 12 euros.

The simulations with FEMM show that we can have a magnetic power increased by 0.2 Tesla (or 2.1 Tesla) compared to the old engine. This geometry makes it possible, among other things, to have a looping length that is twice as small as with a conventional motor.

This is very interesting because it reduces the magnetic leakage losses . With this we can exceed the saturation point of E24 which is usually 1, 9 teslas .

Realization of prototype engines (with a soft iron E24)

Another difficulty in this approach is that it is impossible to radially magnetize a tube-shaped magnet. We were forced to break the tube into sectors 6A, 6B, visible on the Figure 1B , and to magnetize them one by one flat.

A prototype with 5 sectors has been realized. This generates a small loss of motor efficiency compared to a 7-sector system that would be preferred.

Figure 4 appended which consists of Figures 4A (7 sectors) and 4B (5 sectors) and is a top view of Figure 1B .

The average magnetic fields obtained on the prototype motors are 2.05 tesla. This remains a good result since the magnetic field in the gap is greater than the magnetic saturation of the E24.

We were able to obtain a surprising gain on the "tweeter" level of +1.5 dB.

Simulations with 3 three times more (in cm3) of magnet have been realized but the saturation is such that the gain is almost zero.

This suggests that the saturation of E24 does not allow to go further knowing that we can not further focus the magnetic field. This logical way is blocked.

The inventors then had to search for the metal of other materials having a higher magnetic saturation than soft iron (1.9 tesla) and found, among other things, ferrocobalt alloys (49% Co in particular) which possess magnetic saturation of 2.4 Teslas.

Series of magnetic simulations with a ferrocobalt alloy with 49% Co

We performed magnetic simulations with the FEMM ™ software with a motor using as metal this ferrocobalt alloy with heat treatment.

For non-limiting but representative information: a heat treatment at 850 ° C./ for 3 hours makes it possible to increase the permeability by between 1 and 2 teslas relative to an untreated part. This is important because it greatly reduces magnetic leakage and allows it to be concentrated in the air gap. The heat treatment very slightly improves the magnetic saturation value. It goes from 2.3 to 2.4 teslas.

However, it was necessary to limit the parts in ferrocobalt to the strategic places of the engine, for the control of the costs of production.

See attached Figure 5 which consists of Figures 5A through 5C

• champs moyen dans l'entrefer : 2,55 teslas ;
• champs maxi dans le métal : 2,8 teslas ;
• prix du moteur : 200 euros.

Ferrocobalt motor:

• average field in the air gap: 2,55 teslas;
• maxi fields in the metal: 2,8 teslas;
• motor price: 200 euros.

See Figure 6 attached which consists of Figures 6A to 6C

• champs moyen dans l'entrefer : 2,5 teslas ;
• champs maxi dans le métal : 2,8 teslas ;
• prix du moteur : 100 euros.

Engine with only the most domed ferrobalt core:

• average field in the air gap: 2.5 Tesla;
• maxi fields in the metal: 2,8 teslas;
• motor price: 100 euros.

Cf Figure 7 attached which consists of Figures 7A to 7C

• champs moyen dans l'entrefer : 2,42 teslas ;
• champs maxi dans le métal : 2,8 teslas ;
• prix du moteur : 50 euros.

Engine with only the ferrocobalt core:

• average field in the air gap: 2,42 teslas;
• maxi fields in the metal: 2,8 teslas;
• motor price: 50 euros.

Realization of ferrocobalt engine prototypes

We found it interesting to make prototypes for the engine that uses ferrocobalt on the core and on the core plus the dome

This requires the creation of a mold for making blanks.

The performance for the more domed core engine was 2.18 teslas for 2.45 simulated.

Engine performance with only the kernel was 2.18 teslas for 2.45 simulated.

Overall we are far from the simulated results and the difference in performance compared to the E24 is not justified. It was imperative to understand the reason for this anomaly.

Magnetic simulations to simulate possible manufacturing defects: See attached FIG. 8, which consists of FIGS. 8A to 8C.

Addition of a significant mechanical clearance fault between the ferrocobalt dome and the field plate in E24.
Simulated fields without defect: 2.5 teslas.
Actual field measured: 2.18 teslas.
Simulated fields with FeCo / E24 game
0.5mm: 2.45 teslas.

We did not notice a loss of magnetic field. Indeed the area in question is located in an area where the metal is far from saturating (1 tesla). We redid a simulation with a game of 1 mm and the result was unchanged.

See appended FIG. 9 which consists of FIGS. 9A to 9C.

We even pushed the simulations completely removing the piece in E24.
Simulated fields without defect: 2.5 teslas.
Actual field measured: 2.18 teslas.
Simulated fields without the outside room
in E24: 2.35 teslas.

In conclusion, the loss of magnetic field observed on the prototypes does not seem to be related to excessive mechanical clearances or to a problem of connection with the E24.

For information,

See Figure 10 which consists of Figures 10A to 10C

We wanted to estimate the loss of magnetic energy between prototype engines and simulation.

For this we have reduced the amount of magnet used to obtain the same magnetic field as on the prototype engines.

According to the simulation, 60% of magnet must be removed to obtain the same field.

Verification of the energy quality provided by the magnet

For this we simulated the engine with a perfect material (with a magnetic saturation of several thousand teslas). This makes it possible to eliminate losses by magnetic leaks and to check whether there is enough energy supplied by the magnet.

Cf. Figure 11 which consists of Figures 11A to 11C

With a perfect material, the magnetic field in the gap reaches 4.5 Tesla.

We can conclude that there is enough magnetic power provided by the magnet and only the losses related to the magnetic saturation level of the metal can limit the motor performance.

It is possible that the quality of the ferrocobalt used is involved. Either the metal is polluted by non-permeable compounds (carbon, copper, aluminum, etc.) or ferrocobalt is much less than 49% theoretical of the product used, or a heat treatment has destroyed the magnetic characteristics of ferrocobalt.

Other prototypes were then made to find the cause of the problem.

- Prototype with a more powerful neodymium magnet N38M -> N48M

The difference in magnetic field between 2.18 measured and 2.5 simulated is 14%.

An N48M magnet (Br 13.6 kGs) compared to an N38M magnet (Br 12.2 kGs) provides 11% more magnetic power and could compensate for a part of the missing magnetic field.

We measured 2.24 teslas with the N48M for 2.18 teslas with the N38M (first prototype).

There is a small gain but it is much lower than the gain that must provide this more powerful magnet.

This confirms that the problem does not come from the amount of magnetic energy provided by the magnet.

- Prototype with different ferrocobalt processing temperatures

According to the manufacturer's specifications, the temperature of the heat treatment affects the performance of the ferrocobalt.

The maximum performance is obtained at 850 ° C / 3 hours. Above 950 ° C performance begins to fall.

Measurements on prototype engines using parts treated at 850 ° C and 1100 ° C are however substantially identical.

Conclusion:

Despite the probably poor quality of the ferrocobalt supplied and used in the prototypes, the engine using as an external part a soft iron E24 combined with the new arrangement of the engine already offers very good performances since it makes it possible to reach an average field 2, 05 teslas while offering a great decompression.

As an indication, not limiting but representative, the average field measured on the engine of the TGU (in TELAR 57 ™) is 1.95 teslas and on the Focus Ring ™ system (Focal patent EP 02 291 719.9 ) of 1.85 teslas.

This new engine according to the invention can increase the level of the "tweeter" by 1.5 dB compared to the old engine. In addition, it offers a large decompression.

Originality of the chosen solution in terms of performance

From the "tweeter" patented by the Pure Be-inverted Dome Applicator used in the Grande Utopia ™ Be reference chamber, a pure Beryllium-inverted dome tweeter has been redesigned to offer new performance. markedly improved.

The new engine obtained allowed to combine high magnetic performances with a large decompression.

- Increased engine power

This new engine has an average field of 2.05 Tesla. The field increases by 0.2 tesla compared to our previous engine. We managed to overcome the saturation point of the E24. This was possible in another by dividing by 2 length of the smallest loop of the magnetic circuit and tripling the exchange surface between the magnet and the E24 (for 75% more magnet).

See Figure 12 attached

As an indication, the average field measured on the engine of the TGU (in TELAR 57 ™) is 1.90 teslas and on the Focus Ring ™ of 1.85 teslas.

This results in a level increase at the top of the treble of +1.5 dB.

- Increase of the decompression area: See Figure 13 attached:

We went from 500 mm ² to 1155 mm ² of decompression.

This ensures a low resonance frequency between 528 Hz and reduces cavity effects that would cause distortion or dragging

See Figure 14 attached

• 1,75% à 1 kHz contre 1,7% initialement ;
• 0,3% à 2kHz contre 0,5% ;
• 0,17% à 4kHz contre 0,2%.

and have a low distortion rate

• 1.75% at 1 kHz against 1.7% initially;
• 0.3% at 2kHz versus 0.5%;
• 0.17% at 4kHz against 0.2%.

See Figure 15 attached

The "waterfall" shows an attenuation of 40 dB in 0.68 ms over the entire bandwidth with a slower mark at 6 kHz (1 ms).

The "tweeter" (and the new generation engine) according to the invention therefore surprisingly has a higher efficiency despite the absence of mechanical resonance in the useful bandwidth.

This allows to have a dynamic capacity and maintained power increased by 1.5 dB or 41% more noise level while increasing the quality of reproduction of treble.

Such a set of properties is quite interesting and surprising.

General conclusion

• Epaisseur de le membrane : 25 à 50 microns, de préférence 25 microns.
• Extension de la fréquence de coupure haute : 25 kHz - 45 kHz.

On reading the above description and the accompanying drawing, one skilled in the art finds that, with the various tests performed, and the prototype tests, the following results and optimizations are the best compromise. Those skilled in the art will be able to adapt them and make other optimizations by using the above teachings relating to the choice of materials, in particular magnets, the volume of the magnets, the possible "nozzle" effect due to the arrangement and the "Venturi" shape of the core, and / or magnet segments in "cone" or "vertical" tube (the membrane being the horizontal reference), that is to say toward the rear of the engine, information given by the various curves whose type and meaning are all well known to those skilled in the art, without it being possible here to describe all the possible embodiments that are within the direct reach of those skilled in the art:

• Thickness of the membrane: 25 to 50 microns, preferably 25 microns.
• Extension of the high cutoff frequency: 25 kHz - 45 kHz.

Dynamic capacity and power handling increased by 1.5 dB or 41% more noise level while increasing the quality of reproduction of highs.

Membrane very preferably in "inverted dome" that is to say as represented on the Figures 1A 1 C 1D .

Nature of the membrane: very preferably pure beryllium Be.

Magnets magnetized radially arranged in segments around the core, preferably 5 segments, or even better 7 segments, and forming a "vertical" tube or a "rear cone" (as well as polar pieces core and plate field) moved "to the 'rear of the engine', which makes it possible to cleverly clear the compression zone ZC and thus avoid any compression of the air under the membrane; increase in the decompression area, especially to a value around 437 mm 2.

Loop length of the reduced magnetic circuit including a value of 14.7 mm.

Drop in distortion at 2 kHz that drops to 1%.

• augmentation de la surface de décompression de 437 mm² à, dans le prototype, 1100 mm² soit plus du double,
• réduction de la longueur de rebouclage de 14,7 mm à, dans le prototype, 7,75 mm, c'est-à-dire la moitié.
• augmentation du champ magnétique du moteur de 0,2 teslas, qui dépasse donc les 2,05 teslas, et dépasse le point de saturation du fer doux.
• rendement de 95 dB à 98 dB,
• un « tweeter » couvrant 4 octaves, c'est-à-dire à large bande,
• capable d'opérer à pleine puissance ou à plein niveau à partir d'une valeur aussi basse que 2 kHz,
• capable de fournir une puissance élevée,
• le tout avec une très faible distorsion, un très faible traînage ou absence de coloration, une très bonne transparence,
• le tout à un coût de production restant compatible avec les produits très haut de gamme mais néanmoins « grand public » c'est-à-dire pouvant être produit industriellement (et non pas expérimentalement dans seulement un laboratoire comme certains appareils de l'art antérieur).

The best optimization (that is to say optimization AFTER the implementation of the core of the invention, that is to say the new engine, that is to say optimization of the INVENTION and not optimization of the prior art) gives:

• increase of the decompression surface from 437 mm ² to, in the prototype, 1100 mm ² more than double,
• reduction of the looping length from 14.7 mm to, in the prototype, 7.75 mm, that is to say half.
• the 0.2 Tesla motor magnetic field increases, which exceeds 2.05 tesla, and exceeds the saturation point of soft iron.
• efficiency from 95 dB to 98 dB,
• a "tweeter" covering 4 octaves, that is to say broadband,
• capable of operating at full power or at full level from a value as low as 2 kHz,
• able to provide high power,
• all with a very low distortion, very little slack or no color, very good transparency,
• all at a production cost remaining compatible with very high-end products but nevertheless "general public" that is to say that can be produced industrially (and not experimentally in only a laboratory as some devices of the prior art ).

Specific materials and software used Material;

• MLSSA ™ (acoustic measurement system - measurement of response curve and impedance).
• AUDIO PRECISION ™ (acoustic measurement system - distortion measurement).
• DAAS ™ (acoustic measurement system - distortion measurement).
• Gausmeter (magnetic measurement).
• Deaf room.

Software:

• Solidworks ™ (3D drawing soft).
• GEMINI ™ (magnetic simulation).
• FEMM ™ (magnetic simulation).

The invention also covers all the embodiments and all the applications which will be directly accessible to the person skilled in the art upon reading the present application, of his own knowledge, and possibly of simple routine tests.

Claims (20)

Acoustic speaker motor and especially for a "tweeter" transducer "acute", characterized in that the magnet 6 of the motor is transferred "towards the rear of the engine" or "behind the engine" by "pivoting", the magnet 6 becoming aligned with the coil of the transducer, and in that the magnet and pole piece assembly, in particular a field plate 9 and a core 7, is adapted to follow this "pivoting", the field plate assembly, magnet , core, being coaxial or substantially coaxial about the vertical axis of symmetry of the assembly, and generally in the form of a tube or truncated cone "vertical" with possibly a nozzle effect, and possibly with a venturi effect, this by a metal addition at the core, with a "radial" magnetization direction, to clear the ZC volume under the membrane, and thus allow for a larger volume of magnet and / or a larger volume of metal in the pieces polar, note the core 7 and the field plate 9. Engine according to claim 1, characterized in that said transfer and said adaptation make it possible to eliminate or almost eliminate any compression while allowing a larger volume of magnet. Engine according to claim 1, characterized in that instead of adding a larger volume of magnet, more metal is added to the pole pieces. Engine according to Claim 1, characterized in that, thanks to the released volume, an additional volume of magnet and an additional volume of metal in one or the pole pieces are added. Engine according to claim 1, characterized in that said magnet is divided into a plurality of magnet segments 6A, 6B, 6C, ... arranged vertically or at an angle to the vertical, in particular magnets in a vertically collinear position, around 7, and coaxial with the field plate 9, leaving an air gap 2 to accommodate the coil 3. Engine according to claim 2, characterized in that said segments and pole pieces, in particular the core and the field plates, form a "vertical" truncated cone under the membrane with a "nozzle" effect around the core 7. Motor according to any one of claims 1 to 3, characterized in that the nature of the magnet is neodymium, especially N38M, Br 12.2 kGs, or optionally N48M, Br 13.6 kGs. Engine according to any one of claims 1 to 3, characterized in that the pole pieces is a ferrocobalt instead of soft iron optionally heat treated as at 850 ° C / 3 hours and up to 950 ° C. Engine according to any one of claims 1 to 5, characterized in that the metal is a soft iron E24. An engine according to any one of claims 1 to 6, characterized by the following coil: the coil has an inductance of 0.03 mH, an electrical resistance of 6 ohms and a wire diameter of 0.08 mm. Engine according to any one of Claims 1 to 7, characterized in that it is chosen from among the following, with an average field of 2.05 tesla and in particular a field greater than 2 Tesla, which exceeds the saturation point of the iron. sweet: AT the magnet used is neodymium: N38M, the metal used is a soft iron: E24, - length of the smallest circuit loop: 14.7 mm, - exchange surface between the magnet and the E24: 7.8 cm3, amount of magnet: 6.1 cm3; the metal used is a soft iron: E24 length of the smallest circuit loop (magnetic field): 10.82 mm magnet quantity: 6.05 cm3; B - the magnet used is neodymium: N38M - the metal used is a soft iron: E24 - length of the smallest loop of the circuit: 7.75 mm - amount of magnet: 10.5 cm3 - exchange surface between the magnet and the E24: 25.3 cm3; - decompression area is 1 115mm ² ; - average fields in the air gap: 2.1 teslas; - maximum fields in the metal: 2.28 tesla; (surprising gain on the "tweeter" level of +1.5 dB) VS
Ferrocobalt motor: - average fields in the air gap: 2,55 tesla; - maxi fields in the metal: 2,8teslas; D
Engine with only the most domed ferrobalt core: - average fields in the air gap: 2,5 Tesla; - maximum fields in the metal: 2,8 teslas; E
Engine with only the ferrocobalt core: - average fields in the air gap: 2,42 teslas; - maximum fields in the metal: 2,8 teslas; F
Motor with total removal of the soft iron metal part E24;
Fields simulated without defects: 2.5 Tesla;
Actual field measured: 2.18 teslas;
Simulated fields without the outer part in E24: 2,35 teslas. Broadband, that is to say covering 4 octaves, with a high efficiency of 95 dB to 98 dB, allowing full power operation from 1 kHz to 40 kHz, preferably from 2 to 35 kHz, with a very high good power handling, dynamic capacity and power handling increased by 1.5 dB or 41% low distortion, no coloration, gain 25 kHz - 45 kHz, characterized in that it comprises a motor according to the any of claims 1 to 11. "Tweeter" according to claim 12, characterized in that its membrane is pure beryllium (Be). "Tweeter" according to claim 12 or 13, characterized in that the membrane thickness is 2.5 mm to 5.0 mm, preferably 1.5 mm. "Tweeter" according to any one of claims 12 to 14, characterized in that the weight of the membrane is 60 mg. "Tweeter" according to any one of claims 12 to 15, characterized in that its membrane is domed said "inverted". "Tweeter" according to any one of claims 12 to 16, characterized in that its compression zone ZC that is to say the volume between the magnet and under the membrane is released, that is to say say that it lets air circulate including laterally and thus avoids any compression of said air under the membrane. "Tweeter" according to any one of claims 12 to 17, characterized in that the diameter of the decompression surface is reduced to about 50 - 70 and in particular 64 mm2. "Tweeter" according to any one of claims 12 to 18, with a decrease in distortion at 2 kHz to 1%, characterized by : the thickness of the thinned surface by decompression: 1.5 mm, - the thinned surface by decompression reduced to a diameter of 64 mm, - dimensions of the engine mounting brackets have been increased to 8x12 mm - reduction of the decompression area by 437 mm ² . Acoustic speaker characterized in that it comprises a "tweeter" according to any one of claims 12 to 19.

Images