|Publication number||US3821473 A|
|Publication date||Jun 28, 1974|
|Filing date||Nov 10, 1971|
|Priority date||Jun 20, 1969|
|Publication number||US 3821473 A, US 3821473A, US-A-3821473, US3821473 A, US3821473A|
|Original Assignee||Mullins J|
|Export Citation||BiBTeX, EndNote, RefMan|
|Referenced by (28), Classifications (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
United States Patent [191 Mullins June 28, 1974  Inventor: Joe H. Mullins, 2O Oaklawn Rd.,
Fair Haven, NJ. 07701  Filed: Nov. 10, 1971  Appl. No.: 197,405
Related US. Application Data  Continuation-impart of Ser. No. 835,033, June 20,
 US. Cl. 179/1 F  Int. Cl H04r 3/00  Field of Search 179/1 F, 1 FS; 181/31 B  References Cited UNITED STATES PATENTS 1,822,465 9/1931 Trouton 179/] F 1,988,250 l/l935 Olson l81/3l.l
2 I a 5 4. 1+ n 1.. i
6/1962 Koster 179/1 F 3,047,661 7/1962 Winker 179/1 F 3,057,961 10/1962 Turner 179/1 F FOREIGN PATENTS OR APPLICATIONS 659,066 10/1951 Great Britain 179/1 F Primary Examiner-Kathleen H. Claffy Assistant ExaminerJon Bradford Leaheey [5 7] ABSTRACT A High fidelity sound reproduction system comprising an amplifier, an enclosure, a driven speaker mounted in the enclosure and at least one undriven speaker mounted in the enclosure. Each of the speakers mounted in the enclosure have different resonant frequencies and motional feedback devices attached thereto. The outputs of the motional feedback device are combined to produce a negative feedback signal to the amplifier.
5 Claims, 6 Drawing Figures "TENTH- SHEEI 1 0f 2 INVENTOR.
JOE H. MULLINS FIG.3
A TTORNEYS PHENTEBJMZB um SHEETZBfZ FIG.
A TTORNEYS 1 SOUND REPRODUCTION SYSTEM WITH DRIVEN AND UNDRIVEN SPEAKERS AND MOTIONAL FEEDBACK This application is a continuation in part of application Ser. No. 835,033 filed June 20, 1969, now abandoned.
BACKGROUND OF THE INVENTION Extensive research and development in sound input devices, e.g., microphones, tape recorders, phonographs and FM tuners, has resulted in very high fidelity devices. In turn, audio amplifying systems have become sophisticated to a degree whereby amplification of the signal input is achieved with almost completely flat frequency response and very low distortion throughout the audible range.
However, a high fidelity sound reproduction system is only as good as its worst, or lowest fidelity, component, and a necessary part to any sound reproduction system is the acoustical outputtransducer or speaker. By its very nature, this transducer or speaker is usually the largest source of non-uniform response and distortion.
An electro-mechanical speaker, like many mechanical contrivances, has one or more resonant frequencies. If any of these resonant frequencies are in the audible range, an amplified signal input at or near that resonant frequency will result in a higher intensity of sound output than would an identically powered input at a frequency further away from the resonant frequency. This is due to the propensity of the speaker to vibrate at its resonant frequency. A particularly important resonance is .that associated with the lowest frequencies, caused by the combined effects of the speaker mass, the suspension, and usually the air in the closed box surrounding the speaker. It can therefore be seen that the resonances produce a frequency dependent output power. This non-uniform frequency response is undesirable. In addition, large excursions of speakers at low frequencies cause non-linear distortion, which is also undesirable,
When the speaker is put in a closed box, the peaking effects of the resonance can be damped by the inclusion of lossy filling, but the low frequency response of the speaker system falls off rapidly below the resonance.
PRIOR ART These problems have been recognized in the prior art. One solution is to cause negative feedback from the speaker system to alter the characteristics of the signal input to the speaker system. This has been described, for instance, in U.S. Pat. No. 2,860,183 to Ivan Willard Conrad, wherein accelerometers were placed upon the driven element of the speaker.
Since the acceleration of the driven element of a given speaker is proportional to the intensity of the sound produced thereby, at least at low frequencies, if this acceleration is compared electronically with the input to the speaker, the input can be adjusted for constant acoustic power output at all frequencies. In other words, the mechanical system produces an acceleration which is non-uniform with changes in frequency. This non-uniformity is noted by the accelerometer and compared to the amplifier input. The output to the speaker is then adjusted to produce constant acceleration over the full range. It should be noted that this accelerometer feedback method results in no appreciable time delay in the feedback loop, as would the use of an actual microphone used as the sensor. This simplifies the loop stability problem.
Acceleration feedback is capable of producing a system wherein the output sound has a very flat or uniform frequency response. Such systems also automatically produce good transient response. However, the aforementioned accelerometers do not solve a concommitant problem at low frequencies. The excursion of a low frequency speaker is proportional to l/j where f is the frequency. It can be seen that at low frequencies, the excursion is large. It is these large excursions which cause the non-linear distortion referred to above in speakers without the proper feedback. In addition, since these large excursions make low frequency speaker designs difficult and expensive because of the long magnetic field required, any method for reducing the excursion requirement is greatly to be desired.
In the prior art, for example U.S. Pat. No. 1,988,250, passive (non-electrically driven) vibrating elements have been added to speaker system. In the most obvious forms, this would be, for instance, another speaker without a driving coil or a simple portwhere the air in the throat of the port is the undriven element. In the case of an electrically driven speaker (primary) mounted in a closed box with a passive speaker (secondary) mounted in the same box, the motion of the primary speaker causes the air in the box to, in turn, drive the passive speaker. If the low end of the range of the system is close to an appropriate resonance of the secondary, then the excursion of the secondary will be much larger than that of the primary. This is desirable, as large excursions are inexpensive to achieve in the secondary.
It is an object of this invention to provide a novel high fidelity speaker system.
It is a further object of this invention to provide a speaker system that employs speakers having relatively small excursions relative to the low frequency power it produces.
DESCRIPTION OF THE INVENTION The novel speaker system of this invention comprises one or more externally powered driving speakers (primaries) with motional sensors attached to the piston thereof, said speakers being acoustically connected with, and thereby driving, one or more driven speakers (secondaries) with motional sensors attached to the driven portion thereof. This invention is normally applicable to the low frequency portion (woofer) of a multi-speaker system.
Referring now to the drawings:
FIG. 1 is a schematic view of a sound amplification system embodying the principles of the instant invention.
FIG. 2 is a cross-sectional view of a speaker enclosure embodying one of the driven speaker systems of the instant invention.
FIG. 3 is a cross-sectional view of a portion of FIG. 2 showing the driven speakers of the instant invention in plan view.
FIG. 4 is a schematic representation of one embodiment of Amplifier 4 and associated circuitry. v
FIG. 5 is a schematic representation of one embodiment of the circuitry employed to mix and alter the speaker feedback signals to compensate for different speaker areas.
FIG. 6 is a schematic representation of one embodiment of the circuitry employed in the feed-back control unit.
More particularly, FIG. 1. depicts an audio input 2 to an amplifier 4. An amplified signal 5 is conveyed to a sound transducer enclosure 6. The amplified signal then goes to a driving sound'transducer (primary) 16. On the face of the driving transducer 16 is located a motional sensing device 8. The air space within the enclosure 6 is sealed whereby vibrations of the speaker 16 cause alternate rarifications and compressions of the air in the enclosure 6. Also located in the enclosure 6 is a driven sound transducer (secondary) 14 having on its face a motional sensor 10. The driven transducer 14 is powered by the aforesaid alternate rarifications and compressions of the air in the speaker enclosure 6. The signals from the sensors 8 and 10 proceed to a device 12 where their signals are proportionally altered and mixed to account for differences in area between the speakers 16 and 14. The mixed signal 13 is then conveyed to a feedback control device 15 from where it goes to the amplifier 4.
Referring now to FIG. 2. A driving transducer 24 is depicted with one of the motional sensors 25 of the instant invention attached thereto. Elsewhere in the speaker enclosure 7 are two concentrically arranged vibrating surfaces 20 and 22 gasketed from each other and to the speaker enclosure by elastic gaskets 21 and 23. Attached to the vibrating surface 22 is a motional sensor 16. Similarly, a motional sensor 18 is attached to the vibrating surface 20.
FIG. 3 depicts a plan view of the dual driven speaker of FIG. 2.
As shown in FIG. 4, contained within amplifier 4 are circuits which provide high and low frequency compensation to insure that the signal feed back to the input is out of phase with the input signal 2 in a manner known to the art. Out of phase means that for loop gains of slightly greater to slightly less than unity, phase difference between the input signals to the loop and the feed back signal is no less than about 35 degrees.
A circuit to insure this condition at low frequencies is shown in FIG. 4. The values of the components of this circuit will vary, depending on the physical constants (mass, spring constant, damping and resonance) in the driving and driven speakers.
This circuit consists of inputs 2 and feedback signal 31 fed to amplifier 32 in phase opposition, as indicated by plus and minus signs adjacent to amplifier 32. Resistor 33 and capacitor 34. form a low pass network which reduces the loop gain to below unity with the proper phase relationship at high frequencies. The output of this network is fed to driver amplifier 35 and then to output power amplifier 36. The output of amplifier 36 is signal 5 which goes to the driving speaker and is fed back through low frequency network 37 to the input of amplifier 35 at junction 44. This network consists of resistor 38 and 39 in series from signal line 5 to ground. Shunting resistor 39 to ground is a capacitor 40 in series with a tuned network consisting of an inductor 41 and capacitor 42. Connected to the junction between capacitor 40 and inductor 41 is capacitor 43 which feeds the output of network 37 to junction 44, and thus back to amplifier 35. Network 37 serves to reduce the loop gain to below unity with proper phase relationship at low frequency. The components in network 37 are substantially electrical analogs to the physical (mechanical and acoustic) constants of the speaker system, including mass and spring constant of primary and secondary speakers and spring constant of air in the enclosure. The network output forms an electrical response which is equivalent to the inverse of the response of the speaker system at these frequencies.
A signal output proportional to some motional function such as acceleration of the driving speaker times the area of the driving speaker (primary) is added to a signal output identically proportional to the same motional function of the driven speakers times the area of the driven (secondary) speakers. This may be accomplished by adjusting the motional sensor output so as to compensate for the areas of the respective elements. This compensation is necessary because the total power of sound generated is proportional not only to acceleration (at lower frequencies), but to volume of air moved. In turn, the volume of air moved is proportional to the area of the moving surface or speaker. That is, of two speakers sounding a low note at identical acceleration, the larger will produce a greater acoustic power.
The output of the accelerometer can be adjusted to compensate for area by, in the case of a cantilevered piezoresistive accelerometer, adjusting the mass of the cantilever. Alternatively, the output from each speakers accelerometer can be electronically altered in 12 to compensate for differences in the area of the speakers.
As shown by FIG. 5, the signals from the motional sensor 8 and 10 are proportionately altered and mixed by the circuitry referred to by the reference inverse 12 in FIG. 1.
In one embodiment of this circuitry, signals 51 and 52from motional sensors 8 and 10 respectively are fed through resistor 53 and 54 respectively to the input of amplifier 56 at junction 55. The output of amplifier 56 is signal 13 which is fed back, at junction 58, through resistor 57. Resistors 53 and 57 set the gain of the output from sensor 8 and resistors 54 and 57 set the gain of the output from sensor 10. These two signals are summed at the junction point 55.
It should be noted that the values of resistors 53 and 54 are selected so that the ratio of these resistors is inversely proportional to the ratio of the areas of the speakers. These summed signals then are fed back to velocity. In this event, the output of accelerometers may still be used, but should first be electronically integrated.
The sensors can be attached to the piston by simply gluing them, using a stable epoxy resin, for instance. Also, more than one accelerometer can be used on a single speaker. This multiple accelerometer method has the advantage of, when properly arranged, producing a mean acceleration of the moving piston face,
thereby tending to correct error due to breakup or secondary wave generation on the piston face.
Elsewhere in the speaker enclosure may be located one or more driven speakers (secondaries). It is preferred that the secondaries by physically located within approximately one-half wave length of the driving speakers (at the highest usable secondary frequency), or else the operation of the several elements together will produce undesirable directional radiation patterns.
The secondaries can be actual speakers less field coils and magnet structure. as indicated above, or can be of other configurations. For instance, a plug can be taken out of the enclosure, decreased in diameter, and replaced with an elastic gasket surrounding it. In turn, this plug can ha-ve'a plug taken out of it, the smaller plug reduced in diameter, gasketed, and replaced. This results in two concentric driven speakers or secondaries. By this concentric arrangement, more secondary speakers can be placed in a given enclosure wall area.
In case more than one secondary is used, it is necessary that some of the multiple secondaries have mechanical connections to each other or to the primary unless they are tuned to approximately the same'frequency. This is required because below the resonance of the higher frequency secondary, this element is in phase opposition with the driver. In this frequency region, the acoustic output of the primary (driving) speaker and the upper secondary are cancelling. If there is a lower frequency mechanically uncoupled driven element, it will not be driven by the compression and rarifications of the air within the box because the upper frequency secondary will move freely'in response to the driving element and not allow compressions and rarifications to build up in the box. The mechanical connections maybe used to modify this condition. At least one motional sensor is attached to the face of each of the secondaries in a fashion similar to that indicated above.
The reflexive driven elements should be designed such that their resonant frequency is at a point near but below that point where the inefficiency due-to distance below the reconsant frequency becomes unacceptable in the driving speaker. For instance, if a driving speaker has a resonant frequency of about 60 Hertz, for a typical coil resistance-dominated case, it will have an unacceptable output drop of about 12 Db at about 30 Hertz, still in the audible range. Accordingly, the reflexive port (secondary) is tuned so as to have a resonant frequency of about 30 Hertz. Then, when the driving speaker reaches its lower, most inefficient range, the secondary is reaching its most efficient range and pro- 'duces its highest volume of sound. This tends to broaden the frequency range in which there is a flat response, thus decreasing the requirements on the amplifier to provide sufficient power to maintain a constant output power in response to the demands of the feedback condition.
The advantage of multiple secondaries in a mechanically coupled arrangement now becomes apparent in that each of them can be so designed such that its resonant frequency is near that point at which the output drop of the next smaller sized secondary becomes unacceptable. I
Therefore, the chief advantages of the secondary or driven speaker are:
tortion, since very low frequencies are carried by th secondary, and higher frequencies are not.
4. It allows the use of a smaller enclosure (baffle) for a system with a given low frequency limit and a given power amplifier and speaker.
.Above a certain frequency, which frequency is a function of the size and shape of the sound producing speaker, sound power output is no longer proportional to the acceleration of the driving piston, but is rather proportional to the velocity of the driving piston. This point is given for example by the formula d/A E 0.45 for a flat disc speaker in an infinite plane, where a diameter of the speaker )t the wave length of the radiation sound wave. Accordingly, in the feedback loop, a filter network 15 can be provided that notes frequencies above this crossover point, integrates them and feeds them into the signal input amplifier as velocity rather than as acceleration. Design conditions will determine that blend of acceleration and velocity feedback which is most suitable for a given speaker piston shape.
For example, in the above mentioned case of a flat disc in an infinite plane, it can be shown that the sound intensity on the axis of the disc remains constant for all frequencies for constant disc acceleration even though the total power decreases above the above mentioned critical frequency. This is caused by the radiation pattern becoming more directional. Thus, for constant onaxis intensity, the above mentioned integration would not be employed. In general, some combination of the methods would be employed, depending upon the shape of the radiatorand the desired result.
In the case of acceleration feedback, device 15 is eliminated and signal 13 and 31 become the same.
If constand output power is desired, the velocity feedback must be employed above the mentioned critical frequency, f, where f c/)\ and c velocity of sound. However, the system need not be set for constant output power.
The case of velocity feedback is represented by part of FIG. 6. Signal 13 is fed to amplifier 63 through resistor 62. The feedback network,consisting of resistor 64 and capacitor 65 combine with resistor 62 to produce the output 68 which is proportional to velocity above the frequency, F V2 11 Rc where R and 0 refer to ele-,
ments 64 and 65. The remainder of the circuit of FIG. 6 is not used in this case.
It is not required to select either velocity or acceleration feedback alone. When a combination is used, all
of the circuits in FIG. 6 will be employed. Signal 13 is. also fed from junction 61 to variable potentiometer 67 1. In a sound reproduction system comprising an audio signal source, an amplifier for said audio signal source functionally connected with said source, an acoustical output transducer enclosure containing a plurality of acoustical output transducers, the improvement which comprises a system for negative feedback to said amplifier, said system comprising a driving acoustical output transducer responsive to an output of said amplifier and a driven acoustical output transducer responsive to physical vibration of said driving acoustical output transducer, with both driving and driven speakers having motional sensors functionally attached thereto, a first motional sensor functionally attached to said driving acoustical output transducer, a second motional sensor functionally attached to said driven acoustical output transducer, said motional sensors providing electrical signals proportional (to the acoustical power and a motional functions of the acoustical output transducer) to a function of said transducer, said function selected from the class consisting of acoustical power and acoustical intensity which are motional functions of the acoustical output transducers to which they are connected, means for combining said electrical signals and means for employing said combined signals said signals being used as a source of said negative feedback.
2. The reproduction system of claim 1 wherein the driven acoustical output transducer has a resonant frequency in a frequency range where the driving speaker produces a relatively low power of sound for a given power input to it.
3. The reproduction system of claim 1 wherein the driven acoustical output transducer has a resonant frequency below a frequency range where the driving speaker produces a relatively low power of sound for a given power input to it.
4. The reproduction system of claim 1 wherein the driven acoustical output transducer has a resonant-frequency above a frequency range where the driving speaker produces a relatively low power of sound for a given power input to it.
5. The reproduction system of claim 1 wherein the enclosure is substantially airtight and the driven speaker is driven by fluctuations in air pressure in the box caused by the vibrations of said driving speaker.
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US4229619 *||Oct 31, 1978||Oct 21, 1980||Victor Company Of Japan, Limited||Method and apparatus for driving a multi way speaker system|
|US4256923 *||Aug 17, 1979||Mar 17, 1981||Meyers Stanley T||Sound reproducing system utilizing motional feedback and integrated magnetic structure|
|US4276443 *||Aug 17, 1979||Jun 30, 1981||Meyers Stanley T||Sound reproducing system utilizing motional feedback and velocity-frequency equalization|
|US4327250 *||May 3, 1979||Apr 27, 1982||Electro Audio Dynamics Inc.||Dynamic speaker equalizer|
|US4387270 *||Nov 28, 1980||Jun 7, 1983||Tokyo Shibaura Denki Kabushiki Kaisha||Method of harmonizing the impedances of an audio amplifier and loudspeaker interconnected therewith|
|US4550430 *||Feb 20, 1981||Oct 29, 1985||Meyers Stanley T||Sound reproducing system utilizing motional feedback and an improved integrated magnetic structure|
|US4644581 *||Jun 27, 1985||Feb 17, 1987||Bose Corporation||Headphone with sound pressure sensing means|
|US4759065 *||Sep 22, 1986||Jul 19, 1988||Harman International Industries, Incorporated||Automotive sound system|
|US4809338 *||Oct 8, 1986||Feb 28, 1989||Harman International Industries, Incorporated||Automotive sound system|
|US5004066 *||Mar 30, 1989||Apr 2, 1991||Yamaha Corporation||Acoustic apparatus|
|US5012890 *||Mar 15, 1989||May 7, 1991||Yamaha Corporation||Acoustic apparatus|
|US5173575 *||Mar 21, 1989||Dec 22, 1992||Yamaha Corporation||Acoustic apparatus|
|US5418873 *||Sep 9, 1993||May 23, 1995||Digisonix, Inc.||Active acoustic attenuation system with indirect error sensing|
|US5471527||Dec 2, 1993||Nov 28, 1995||Dsc Communications Corporation||Voice enhancement system and method|
|US5533134 *||Aug 8, 1994||Jul 2, 1996||Sony Corporation||Motional feedback loudspeaker apparatus having a coupling member for connecting a voice coil bobbin with a detecting coil bobbin|
|US5588065 *||May 23, 1995||Dec 24, 1996||Masushita Electric Industrial Co.||Bass reproduction speaker apparatus|
|US5649015 *||Sep 15, 1995||Jul 15, 1997||Midnite Kitty, Inc.||Speaker simulator|
|US7113607||Sep 2, 1999||Sep 26, 2006||Mullins Joe H||Low frequency feedback controlled audio system|
|US7260229||Oct 16, 2002||Aug 21, 2007||Audio Products International Corp.||Position sensor for a loudspeaker|
|US7961892||Jul 28, 2003||Jun 14, 2011||Texas Instruments Incorporated||Apparatus and method for monitoring speaker cone displacement in an audio speaker|
|US8401207||Mar 26, 2010||Mar 19, 2013||Harman International Industries, Incorporated||Motional feedback system|
|US20030072462 *||Oct 15, 2002||Apr 17, 2003||Hlibowicki Stefan R.||Loudspeaker with large displacement motional feedback|
|US20030086576 *||Oct 16, 2002||May 8, 2003||Hlibowicki Stefan R||Position sensor for a loudspeaker|
|US20050025317 *||Jul 28, 2003||Feb 3, 2005||Fedigan Stephen John||Apparatus and method for monitoring speaker cone displacement in an audio speaker|
|US20090060213 *||Jan 18, 2007||Mar 5, 2009||Harry Bachmann||Method for Determining the Position of a Moving Part in an Electroacoustic Transducer|
|DE102007002920A1||Jan 19, 2007||Jul 31, 2008||Baumann, Dieter, Dipl.-Ing.||Device for use in loudspeakers, has moving coils and diaphragm speed is determined with help of moving coil or together with attached additional coil|
|EP0340435A2 *||Mar 20, 1989||Nov 8, 1989||Yamaha Corporation||Acoustic apparatus|
|EP0548836A1 *||Dec 18, 1992||Jun 30, 1993||Matsushita Electric Industrial Co., Ltd.||A bass reproduction speaker apparatus|