GB2268356A - High-fidelity loudspeaker. - Google Patents

Abstract

A integral high-fidelity loudspeaker, comprising a primary transducer element 10 for transducing a primary electrical signal 12 to a mechanical vibration and a sound producing element 11 coupled to the primary transducer element and responsive to the mechanical vibration thereof for producing sound. A secondary transducer element 14 is coupled to the sound producing element for producing a secondary electrical signal proportional to a variation in position of the sound producing element A feedback circuit 17, 19 has an output connected to an input of the primary transducer element and has a pair of inputs respectively coupled to the primary and secondary electrical signals for producing at its output an error signal 18 proportional to a difference between the primary and secondary electrical signals. The primary transducer element is thus servo-bound to the primary electrical signal and follows it faithfully, substantially independently of its mechanical characteristics. The loudspeaker element may be provided as an integral unit which does not require any special matching to the amplifier to which it is coupled and therefore is completely portable between different high-fidelity systems. <IMAGE>

Description

High-Fidelity Loudspeaker This invention relates to a high-fidelity loudspeaker.

Notwithstanding the great advances which have been made in the art of sound production and reproduction, it is a well-known fact that the weak link of all such systems is the loudspeaker. A "highfidelity" sound system is one where the loudspeaker produces sound which is a faithful reproduction of the originating sound source. As an ideal criterion, a blind-folded listener should not be able to distinguish between an original sound, be it music or speech, and the same sound after reproduction by a "high-fidelity" sound system.

So far as the loudspeaker element itself is concerned, a "high-fidelity" loudspeaker is one wherein the movement of the sound producing element matches the shape of the electrical signal fed thereto and is proportional to its magnitude for the whole audio range of frequencies.

The present invention does not relate to external means for improving the. sound quality of a loudspeaker such as, for example, baffles but is concerned with the electro-mechanical transducer element which converts an electrical signal into sound.

Improvements to high-fidelity loudspeakers have been achieved along two main lines: in the first instance, by careful design of its mechanical characteristics, the proper or resonant frequency of the loudspeaker can be reduced or driven outside the audio range. Additional improvements can be obtained by the association of two or more loudspeakers of different characteristics. In order to obtain the required flatness of response for the whole range of audio frequencies, it may also be necessary to connect bulky filter elements, chokes and capacitors to the loudspeaker terminals. Such improvements as have been made have inevitably resulted in the high-fidelity loudspeaker being very expensive.

Additionally, since such improvements as have been effected are dependent on the mechanical characteristics of the loudspeaker, they have inevitably only been partial. Thus, the flutter of the paper cone associated with electrodynamic loudspeakers is difficult to eliminate without increasing the mass of the sound-producing element. However, doing this clearly alters the mechanical properties of the resulting loudspeaker and thus defeats other improvements incorporated within the mechanical design.

It is also known to improve the frequency response of a loudspeaker by incorporating a feedback system to the main amplifier of the sound reproducing system. Such a feedback system is responsive to the actual movement of the loudspeaker's sound-producing element, for producing an error signal corresponding to the difference between the actual movement and the desired movement of the loudspeaker's soundproducing element. This permits servo control of the loudspeaker's sound-producing element, whereby very high-fidelity sound reproduction may be achieved.

Thus, for example, U.S. Patent No. 3,014,096 (Clements) discloses a sound reproducing means including an excursion gauge transducer which develops an electrical signal whose amplitude is instantaneously proportional to the departure from a given mean position of the sound reproducing element of the loudspeaker. The excursion gauge transducer is basically a velocity meter whose output is integrated so that the resulting, integrated signal is proportional to displacement of the loudspeaker cone. Theoretically, the signal developed by the excursion gauge transducer ought to be identical to the primary signal within the voice coil of the loudspeaker. Thus, any discrepancy between the primary signal developed by the voice coil and the secondary signal developed by the excursion gauge transducer represents an error in the fidelity of the loudspeaker movement.The error signal is fed to an amplifier which is external to the loudspeaker itself and typically is constituted by a high-fidelity amplifier within a sound reproduction system. By such connection, the feedback loop is closed and the loudspeaker cone faithfully follows the primary electrical signal.

The above principle is also disclosed in U.S. Patent No.

1,822,758 (Toulon) wherein an auxiliary coil is employed for producing an EMF which should be exactly equal to a corresponding signal furnished by a microphone adjacent the loudspeaker cone for developing a signal proportional to a displacement of the loudspeaker cone.

It will be clear that coupling the loudspeaker element to the amplifier in the manner proposed by the above-mentioned patents requires means to correct any phase alteration due to the conventional use of transformer and capacitive coupling between successive amplifier stages.

It is also known to employ filters for improving the frequency response of the system. Thus, for example, U.S. Patent No.

2,857,461 (Brodie) employs a capacitor-resistor network functioning as a filter for coupling the output generated by the secondary transducer to the input of an amplifier to which the loudspeaker is connected.

The amplifier in such high-fidelity systems generally comprises several stages connected in cascade, with successive stages being mutually coupled through transformers or capacitors. Generally, the purpose of such coupling is to pass the small a.c. signal only whilst blocking any d.c. component in order that subsequent stages of amplification are not driven into saturation.

Such coupling as well as the provision of filters for coupling the secondary transducer to the input of the amplifier all point to the need for careful adjustment of the system in order to compensate for any phase alteration and to allow the system to function properly without oscillating. Furthermore, even if such systems achieve a faithful reproduction of musical sounds which are harmonic in nature, they may not be able to reproduce exactly short pulses, noise, or other non-periodic sounds such as may occur in speech consonants or in the attack of some musical instruments.

The use of negative feedback for improving the fidelity of a loudspeaker was proposed by Toulon (referred to above) as early as 1929. It will be clear from what is stated above that the general concept has been well developed in order to improve the fidelity of the soundreproduction system. All this, notwithstanding, it is apparent that highfidelity loudspeakers of the kind envisaged by the invention are not available. In order to understand why this might be, even apart from the critical adjustment of the feedback circuit which requires high tolerance components and manual adjustment and is therefore expensive, it is also believed that a major contributing factor is the development of highfidelity systems themselves.

Thus, in high-fidelity systems, the loudspeakers are almost invariably sold as separate units to be coupled, by the listener, to the amplifier of the high-fidelity system. Typically, in such high-fidelity systems, the loudspeakers may well be produced by a different manufacturer to that of the amplifier itself. Indeed, it is by no means uncommon for all of the elements in a high-fidelity system to be produced by different manufacturers. This is hardly surprising since the manufacture of high-fidelity equipment is highly specialized and manufacturers have therefore tended to concentrate in one particular area of expertise.

Consequently, none of the prior art high-fidelity loudspeakers is susceptible to use in such systems owing to the requirement to couple the output from the feedback circuit to the input of the main amplifier. Clearly, this can be done only when the main amplifier and the loudspeaker are provided as an integral unit. Of course, strictly speaking, in the case where the amplifier and the loudspeaker are produced by the same manufacturer, the feedback coupling circuitry could indeed be provided within the amplifier and connected to a terminal within the amplifier casing, so as to allow connection thereof to the secondary transducer within the loudspeaker. In practice, however, this is not done, possibly because the additional components merely add to the price of the amplifier whilst rendering the loudspeaker completely unsuitable for use with other amplification systems.This is a very serious drawback in an art where interchangeability of system elements is a prime requirement.

It is an object of the invention to provide a high-fidelity loudspeaker element which does not require critical mechanical characteristics, nor careful electronic adjustment and which can reproduce nonharmonic signals faithfully and can be isolated as a single high-fidelity element of the sound reproduction chain.

According to a broad aspect of the invention there is provided a high-fidelity loudspeaker, comprising: a primary transducer element for transducing a primary electrical signal to a mechanical vibration, a sound producing element coupled to the primary transducer element and responsive to the mechanical vibration thereof for producing sound, a secondary transducer element coupled to the sound producing element for producing a secondary electrical signal proportional to a variation in position of the sound producing element, and a feedback circuit including a differential amplifier coupled to a power amplifier and having an output connected to an input of the primary transducer element, said differential amplifier having a pair of inputs respectively coupled to the primary and secondary electrical signals for producing at said output an error signal proportional to a difference between the primary and secondary electrical signals; such that when the loudspeaker is employed within a high-fidelity system comprising an amplifier of predetermined characteristics, the loudspeaker is completely independent of the characteristics of the amplifier.

The feedback circuit is free of any coupling element which might alter the phase of the resulting signal. This condition may be satisfied, for instance, by the use of solid state amplifiers and resistive coupling, the final d.c. level of the amplifier chain be stabilized by a negative feedback of appropriately long time constant.

By means of such a feedback circuit, a closed loop having zero phase alteration is formed, in which the primary transducer element is enslaved to the primary electrical signal by an amplified signal which does not permit the resulting secondary electrical signal to differ therefrom by any significant amount.

Since the power amplifier is already provided as an integral part of the loudspeaker, the main amplifier in the high-fidelity system need not be followed by a power amplifier. This renders the loudspeaker element according to the invention independent of the characteristics of the main amplifier.

The primary transducer element may be any conventional loudspeaker element such as, for example, electromagnetic, electrodynamic and piezo-electric types of speaker. In a preferred embodiment of the invention, the primary and secondary transducer elements are moving coils connected to the loudspeaker cone, whose movement relative to a magnetic field induces an electrical voltage in the secondary coil. In such an arrangement, the magnitude of the electrical voltage is proportional to the time differential of the coil's movement and an integrator is therefore provided so as to produce a signal whose magnitude is proportional to the varying distance of the coil from its rest position.

For a clearer understanding of the invention and to see how the same may be carried out in practice, a preferred embodiment will now be described with reference to the accompanying drawings, in which: Figs. la and Ib are schematic representations of a prior art loudspeaker element and of a feedback circuit incorporating the loudspeaker element; Fig. 2 shows schematically a high-fidelity loudspeaker according to the invention employing the loudspeaker element shown in Fig. 1; Fig. 3 shows schematically the principal components of a movingcoil type loudspeaker element; Fig. 4 shows a detail of the arrangement depicted in Fig. 3 and further including a secondary transducer element in accordance with the invention; Fig. 5 shows typical waveforms associated with the arrangement shown in Fig. 2;; Fig. 6 is an electrical circuit diagram of a preferred embodiment of the invention employing the loudspeaker element shown schematically in Fig. 4; Fig. 7 shows experimentally derived frequency response curves for the embodiment shown in Fig. 6 and for the primary loudspeaker element shown in Fig. 3; and Fig. 8 show pictorial representations of the mechanical response of a loudspeaker element in comparison to an electrical input signal both with and without the provision of the feedback circuit shown in Fig. 6.

Fig. la shows a loudspeaker element 10 (constituting a primary transducer element) including a cone 11 for producing sound in response to an audio-frequency signal 12 (constituting a primary electrical signal) fed to an input thereof. The sound produced by the vibration of the moving cone 11 approximately matches the characteristics of the electrical signal 12, distortions being essentially due to the inherent mechanical characteristics of the loudspeaker element 10.

Fig. 1b shows schematically a typical prior art feedback arrangement incorporating the loudspeaker element 10 depicted in Fig. 1a and as described in the patents referred to above. In such an arrangement, the electrical signal 12 is fed to a first input of an amplifier 13 whose output is connected to the loudspeaker element 10. Also provided is a secondary transducer element 14 which is responsive to the mechanical movement of the cone 11 for producing a secondary electrical signal representative of the magnitude of the mechanical vibration of the cone 11. An output of the secondary transducer element 14 is fed back, via a filter 15 to a second input of the amplifier 13.The filter 15 must be carefully matched to the characteristics of the loudspeaker element 10 and the amplifier 13 and thus militates against the use of an amplifier and loudspeaker element produced by different manufacturers.

Fig. 2 shows a general scheme of a system according to the invention for reducing the distortions associated with the loudspeaker element 10. Coupled to the loudspeaker element 10 is a secondary transducer element 15 which is responsive to the mechanical movement of the cone 11 for- producing.a secondary electrical signal 16 representative of the magnitude of the mechanical vibration of the cone 11. Both the primary and secondary electrical signals 12 and 16, respectively, are fed to a differential amplifier 17 which produces a differential signal 18 corresponding to the error between the primary and secondary electrical signals. The error signal 18 is amplified by a current amplifier 19 and then fed back to the loudspeaker element 10.

The operation of such an arrangement is as follows. On first applying the primary electrical signal 12 to the arrangement, before the cone 11 of the loudspeaker element 10 moves, the secondary electrical signal 16 will be zero. Consequently, the error signal 18 fed back to the loudspeaker element 10 corresponds to the primary electrical signal 12 whose reproduction by the loudspeaker element 10 is required. When the movement of the cone 11 faithfully reproduces the primary electrical signal 12, the error signal 18 will be zero, no correction signal thus being applied to the loudspeaker element 10. The loudspeaker element 10 is thus servo-bound to the primary electrical signal 12 and follows it faithfully, substantially independently of its mechanical characteristics, within reasonable limits.

Fig. 3 shows the main components of a high-fidelity loudspeaker according to the principles described above with reference to Fig. 2 and employing a conventional electrodynamic loudspeaker having a cardboard cone 20 fixed to a rigid frame 21 via a flexible coupling 22. A coil 24 fixed to the cone 20 is free to move axially relative to a magnetic field produced by a strong magnet 25. Such movement is achieved by feeding an electrical signal 26 to the coil 24 via two leads 27 and 28 attached to opposite ends of the coil 24 so that the changing current applied to the coil 24 interacts with the magnetic field of the magnet 25 so as to produce axial movement of the coil 24 and hence vibration of the cone 20 attached thereto.

Among numerous details affecting the sound quality of such an arrangement, the mass of the cone 20 together with the components fixed thereto and the compliance of the flexible attachment 22 determine the frequency characteristics of the loudspeaker, while the magnetic field intensity determines its damping effectiveness on the coil.

Fig. 4 shows how such an arrangement may be adapted in accordance with the invention by the addition of a second coil 30 (constituting a secondary transducer element) rigidly attached to the cone 20 so as to execute axial movement in the magnetic field of the magnet 25 owing to movement of the cone 20. Such axial movement induces within the coil 30 a voltage whose magnitude is proportional to the time differential of the movement of the coil 30 and thus of the cone 20 attached thereto. Consequently, the magnitude of the voltage induced within the coil 30 and represented by a waveform 31 is a function of the velocity of the cone 20 whose movement is represented by a waveform 32 in the Figure.Specifically, an integrator 33 connected across the coil 30 integrates the electrical signal 31 induced therein so as to produce the secondary electrical signal 16 whose magnitude is proportional to the movement of the secondary coil 30 and thus of the cone 20 attached thereto.

In such an arrangement the secondary coil 30 together with the integrator 33 constitute the secondary transducer element 15 shown in Fig. 2.

Referring now to Fig. 5 of the drawings there are shown typical waveforms associated with the arrangement of Fig. 2 employing the secondary transducer element shown in Fig. 4. Thus, reading from left to right, there are shown pictorially waveforms of the primary electrical signal 12, the movement 32 of the cone 20, the resulting output signal 31 from the secondary coil 30 as fed to the integrator 33 and the secondary electrical signal 16 produced thereby reproducing electrically the movement 32 of the cone 20.

Fig. 6 shows schematically a circuit diagram for implementing the arrangement shown in Fig. 2 and employing an electrodynamic loudspeaker element 35. Associated with the loudspeaker element 35 is a cone (not shown) to which there are coupled primary and secondary coils (not shown) having respective input leads 36, 37 and output leads 38, 39. The corresponding input lead 37 and output lead 39 of the primary and secondary coils are connected to ground (GND), the remaining output lead 38 of the secondary coil being connected via a resistor 40 to a first input 41 of an operational amplifier 42 whose second input 43 is connected to GND via a resistor 44. Connected across an output 45 of the operational amplifier 42 and its first input 41 is a capacitor 47 and a resistor 48.

The output 45 of the operational amplifier 42 is connected via a resistor 50 to a first input 51 of an operational amplifier 52 whose second input 53 is connected to GND via a resistor 54 in series with a capacitor 55.

An output 58 of the operational amplifier 52 is coupled to the first input 51 thereof via a resistor 60 and is likewise coupled to a first input 61 of an operational amplifier 62 via a coupling resistor 63.

Connected across the first input 61 of the operational amplifier 62 and an output 65 thereof is a resistor 66. The output 65 is connected to the junction of the resistor 54 and the capacitor 55 via a resistor 67. The output 65 of the operational amplifier 62 is also connected via a dioderesistor biasing network depicted generally as 68 to an input of a power amplifier depicted generally as 69 and including two current amplifiers 70 and 71 connected in push-pull. The current amplifier 70 comprises an NPN darlington whilst the current amplifier 71 comprises a PNP darlington. An output 72 of the power amplifier 69 is fed back to the input lead 36 of the primary coil of the loudspeaker element 35.

The diodes in the resistor-diode network 68 are low current rectifier diodes having a reverse bias voltage VBE equal to approximately 0.7 V. Two diodes on each side of the input to the power amplifier 69 together with the respective resistors provide the necessary bias to the emitters of the darlingtons 70 and 71. This ensures that the darlingtons start to conduct as soon as an input voltage is applied thereto and thus avoids discontinuities between the positive and negative sides of the output current.

An audio-frequency electrical signal 75 is fed between a second input 76 of the operational amplifier 62 via a resistor 77 and GND, the signal 75 corresponding to the primary electrical signal 12 shown in Fig. 2 and which is to be faithfully reproduced acoustically by the loudspeaker element 35.

Also shown in Fig. 6 are equipotential supply points denoted as A and B which are typically +15 V and -15 V, respectively and may be derived from suitable batteries 79 and 80 or, more likely, a suitable filtered d.c. power supply.

The operation of the circuit shown in Fig. 6 is as follows.

The operational amplifier 42 in conjunction with the capacitor 47 and the resistor 48 constitutes the integrator 33 shown in Fig. 4. The operational amplifier 52 together with its associated circuitry is a pre-amplifier which amplifies the output of the integrator and feeds the amplified signal to the operational amplifier 62 which, together with its associated circuitry, constitutes the differential amplifier 17 shown in Fig. 2. The integrated signal appearing at the first input 61 of the differential amplifier 62 is proportional to the distance from its mean position of the secondary coil and thus of the cone of the loudspeaker element 35. The primary electrical signal 75 is fed to the second input 76 of the operational amplifier 62.

Consequently, the output 65 of the differential amplifier 62 which, after amplification by the power amplifier 69, is fed back to the primary coil input 36, is proportional to the difference between the actual movement of the loudspeaker element 35 and the desired movement represented by the waveform 75. By this means, the primary coil of the loudspeaker element 35 is servo-controlled so that the sound producing cone coupled thereto faithfully follows the primary electrical signal 75.

In the arrangement shown in Fig. 6, the value of the capacitor 47 need not be very high since the integrator 32 need be responsive only to signals in the audio range of frequencies. Typically, a value of 2 pF is sufficient. The coupling between the output 38 of the secondary coil and the input 36 of the primary coil via the complete chain of amplifiers 42, 52, 62 and 69 must not introduce any phase difference to the signal derived by the secondary coil in order that the feedback from the output 72 of the power amplifier 69 to the input 36 of the primary coil may be correctly effective. The simplest way to eliminate introduction of a phase difference between the two signals is to avoid capacitive coupling and rely only on direct d.c. coupling via the resistors 40, 44, 50, 54, 63 and 77 all, typically, having a value of 1.2 Kn.

However, in practice, unless very precise and stable components are used, d.c. coupling will amplify to unacceptable levels the slightest d.c. drift generated by the front-end components.

In the circuit shown in Fig. 6, two steps are taken to avoid such undesired amplification. First, the resistor 48 is connected across the capacitor 47 of the integrator so as to fix the amplification factor thereof and thus reduce the output d.c. voltage at zero input. Typically, the resistor 48 has a value of 2 Mn or higher which does not affect the performance of the integrator for the limited range of frequencies over which it must operate and yet reduces the output voltage to a few millivolts for an operational amplifier 42 having reasonable performance.

Secondly, a negative feedback having a large time constant is provided so as to correct any random or drift d.c. voltage. Such feedback is realized via the series combination of the resistor 67 having a value of 10 Mn and the capacitor 55 having a value of 2.2 ,uF. The resulting large time constant avoids any interference from the feedback to the audio frequencies handled by the circuit.

The amplification of the operational amplifier 52 should be adjusted by way of the load resistor 60 so as to obtain approximately the same amplitude as would be achieved when the signal 75 is fed to the loudspeaker element 35 in an open loop provided with power amplification. This is important because, in the closed loop, the output 58 of the operational amplifier 52 is compared at the first input 61 of the differential amplifier 62 to the electrical signal 75 applied to its second input 76.

Thus, the average response amplitude of the loudspeaker element 35 in the closed loop system shown in Fig. 6 will be approximately the same as when fed by the primary electrical signal 75. This constraint applies when the primary electrical signal 75 can be modified as required in order to control the loudspeaker volume. In such case, it was found that an amplification factor of 55 was required.

Alternatively, the amplitude of the primary electrical signal 75 can be fixed and the volume can be controlled by varying the value of the resistor 60.

The operational amplifier 52 should have a flat frequency response characteristic in all of the audio range and can, if necessary, be replaced by more than one operational amplifier connected in cascade.

The function of the operational amplifier 62 and its associated components is to derive the differential signal between the signals appearing at the inputs 61 and 76 thereof and to amplify it so as to render effective the subsequent feedback. However, the gain of the operational amplifier 62 and associated circuitry must be constrained in order to prevent oscillations and other instabilities. In an actual embodiment of the invention reduced to practice, a gain of 4 was found to be satisfactory.

Owing to the high amplification between the integrator and the output of the power amplifier 69 and the high current flowing therein, great care must be taken to avoid spurious feedbacks which would otherwise de-stabilize the circuit. In particular, the ground lines associated with the integrator and associated amplifiers, on the one hand, and the power amplifier 69, on the other hand, must be completely separated and commonly connected to only a single point, GND.

It should also be noted that the input leads 36 and 37 of the primary coil must be correctly connected with respect to the output leads 38 and 39 of the secondary coil in order to achieve negative feedback. If the sense of the leads 36, 37 relative to the leads 38, 39 is inverted, positive feedback will result, blocking the feedback loop to saturation in the positive or negative direction. The correct sense of the leads 36, 37 relative to the leads 38, 39 is easily determined empirically.

Finally, referring to Fig. 7 of the drawings, there are shown experimental frequency response curves for a variable frequency primary electrical signal being applied to a cheap 4" loudspeaker (curve I) and to the identical loudspeaker element fitted with a secondary coil and incorporated within the circuit of Fig. 6 (curve II).

The uncontrolled response curve I shows a peak of approximately 24 dB at a frequency of 117 Hz in addition to other irregularities, while the response of the controlled loudspeaker element (curve II) is substantially flat (to within + 1 dB) for all practical purposes.

In Fig. 8 there are shown three primary electrical signals (a) having respective frequencies of 33, 110 and 988 Hz, together with the corresponding response of the loudspeaker cone when its coil is connected directly to the primary electrical signal (b) and, in (c), when connected via the feedback circuit described above in relation to Fig. 6 of the drawings. While the mechanical response shown in (b) is highly distorted near the resonant frequency of the cone, the response shown in (c) indicates that it quite faithfully reproduces the primary electrical signal when the loudspeaker is coupled through the feedback circuit.

It will be understood that, whilst the invention has been described with particular reference to an electrodynamic, moving-coil type of loudspeaker element, the invention is not limited to the type of primary transducer element employed by the loudspeaker. Thus, the invention can be equally well applied to electromagnetic speakers (as in telephones), piezo-electric loudspeakers etc. Similarly, the secondary transducer employed by the invention is not limited to the use of a coil in a magnetic field but equally well envisages the use of magneto-electric pickups (as used in some gramophones), piezo-electric pickups (such as ceramic pickups used in gramophones), condenser pickups (as used in some microphones), opto-electrical pickups and so on.

It will also be appreciated that whilst the preferred embodiment employs a cone as the sound producing element, this also is not required and any other type of sound producing element may equally well be employed.

In the preferred embodiment, the signal derived by the secondary transducer element, being proportional to the speed of its movement, is integrated in order to produce a signal proportional to the variation in position of the loudspeaker cone. However, the integration must be omitted if the secondary transducer inherently produces such a signal. In practice, such a signal can be realized using opto-electrical means, such as an optical linear encoder coupled to the sound producing element, or a Hall effect detector as described in U.S. Patent No. 4,821,328 (Drozdowski), thereby obviating the need for subsequent integration.

Finally, it should be noted that it is obviously not necessary to incorporate the feedback circuit (represented by elements 17 and 19 in Fig. 2) within the loudspeaker itself and, consequently, a modified loudspeaker element according to the invention merely comprises a conventional loudspeaker having a primary transducer element and a sound producing element and further including a secondary transducer element for producing a signal responsive to the motion of the sound producing element, while the necessary feedback circuit described in detail above can be completely external to such an arrangement.

Likewise, the integrator also, when required, and constituting part of the secondary transducer, need not be mounted proximate the secondary transducer and can be located outside of the loudspeaker.

Claims (7)

CLAIMS:

1. A high-fidelity loudspeaker, comprising: a primary transducer element for transducing a primary electrical signal to a mechanical vibration, a sound producing element coupled to the primary transducer element and responsive to the mechanical vibration thereof for producing sound, a secondary transducer element coupled to the sound producing element for producing a secondary electrical signal proportional to a variation in position of the sound producing element, and a feedback circuit including a differential amplifier coupled to a power amplifier and having an output connected to an input of the primary transducer element, said differential amplifier having a pair of inputs respectively coupled to the primary and secondary electrical signals for producing at said output an error signal proportional to a difference between the primary and secondary electrical signals;; such that when the loudspeaker is employed within a high-fidelity system comprising an amplifier of predetermined characteristics, the loudspeaker is completely independent of the characteristics of the amplifier.

2. The high-fidelity loudspeaker according to Claim 1, wherein the primary transducer element is a moving coil.

3. The high-fidelity loudspeaker according to Claim 1 or 2, wherein the secondary transducer element includes: a moving coil coupled to the sound producing element so that movement thereof relative to a magnetic field induces in said moving coil a voltage proportional to a speed of said movement, and an integrator coupled to the moving coil and responsive to said voltage for producing said secondary electrical signal at an output thereof.

4. The high-fidelity loudspeaker according to Claim 3, further including a voltage amplifier coupled between-the differential amplifier and the primary transducer element.

5. The high-fidelity loudspeaker according to Claim 4, further including a negative feedback having a long time constant for producing a stable d.c. output level of the feedback circuit when the secondary electrical signal input thereto is of zero amplitude.

6. The high-fidelity loudspeaker according to any one of the preceding claims, wherein the sound producing element is a cone.

7. A high-fidelity loudspeaker substantially as described herein with reference to the accompanying drawings.