US3564163A

US3564163A - Ribbon loudspeaker

Info

Publication number: US3564163A
Application number: US632422A
Authority: US
Inventors: Gilbert L Hobrough
Original assignee: Individual
Current assignee: Individual
Priority date: 1967-04-20
Filing date: 1967-04-20
Publication date: 1971-02-16
Anticipated expiration: 1988-02-16

Abstract

This disclosure describes a ribbon-type loudspeaker having a ribbon carrying voice currents and being positioned between magnetic poles. Various means to attenuate or eliminate standing waves on the ribbon are shown.

Description

United States Patent 3,564,163 [72] Inventor svillaert Ligobmugh 5 References Cited UNITED ST TE PA N [2]] App]. No. 632,422 A S TE TS [221 Filed Aim 20,1967 2,164,157 6/1939 Kennedy ..179/138(VEL) [45] Patented Feb 16,1971 2,608,265 8/1952 Eckardt 181/32 [73] Assignee Robert L. Wathams FOREIGN PATENTS a fractional part interest 32,649 France ..179/138(VEL) 609,853 4/1927 France 179/138(VEL) 604,908 7/1948 Great Britain ..179/138(VEL) '[54] RIBBON LOUDSPEAKER Primary Examiner-Ralph D. Blakeslee 16 Claims, 6 Drawing Figs Attorney-Robert L. Nathans s2 U.S. Cl 179/115; I

179/138; 181/31 ABSTRACT: This disclosure describes a ribbon type loud- [51] Int. CL H041 9/00 speaker having a ribbon carrying voice currents and being [50] Field of Search t. 17 9 138 positioned between magnetic poles.

(VEL), 117, 180, 181, 1 19 (R), (Inguired), 115.5V I Various means to attenuate or eliminate standing waves on 181/31, 32 the ribbon are shown.

PRIOR ART V/SCOUS Sig/SOHO MATERIAL 2 211+ .RI QNL U SBEAKER BACKGROUND OF'THETINVENTION Thisinvention relates to the conversion of electrical energy into sound oracoustical energy and-in particular to the improvement of an electromagnetic device for reproducing sound with greater-fidelity than may be achieved by other means.

lnthe reproduction of sound by electrical means,thr ee elements at least are necessary: firstly, an acoustical/electrical transducer or microphone for converting sound energy into electrical energy; secondly, an electrical amplifying means for increasing the strength or intensity of the electrical signals fromthe microphone to the required power level; and lastly,

an electrical acoustical transducer or loudspeaker for converting theamplified electrical signals representing theoriginal sound back into acoustical energy. 7

Microphones and electrical amplifying means have reached a state of development such that they may introduce negligible distortion or alteration in the quality of the reproduced sound. The design of loudspeakers, however, has not reached this level of performance. At the present time, loudspeakers are limited at the low frequency end of the sound spectrum by the requirement for either a very large size or alternatively for a low efficiency of transduction. Atthe high frequency end of the sound spectrum, loudspeaker performance is seriously impaired by the strong tendency of the vibrating and radiating surfaces of a loudspeaker diaphragm to develop complex resonance patterns in which partsof such surfaces vibrate at high amplitude at certain frequencies and independently or in opposition to vibrations occuringin other parts thereof.

In the mid or speech portion .of the spectrum, speaker design is complicated by the greatly enhanced ability of the human ear to discriminate and sensedistortions. In particular transient distortion such as that induced by diaphragm mode resonances, are especially disturbing to the ear and interfere seriously with speech recognition and musical enjoyment.

Transient distortion arising out of resonant modes is one of the most troublesome problems in loudspeaker design. Most speaker distortions are alleviated by a-reduction in frequency range to be reproduced. Accordingly it is a common practice in high fidelity sound reproduction systems to employ several loudspeakers to cover the sound spectrum, each speaker being restricted by a filter network to a fraction of the-total spectrum. By this means a speaker can be designed to operate most effectively in one part of the'spectrum, the filter network blocking those parts of the spectrum in which speaker performance has been compromised. For example, a large diaphragm speaker for low frequency operation would suffer from mode resonances or cone breakup in the midand high frequency ranges. A low pass-filter would eliminate the mid and high frequencies from the speaker thus avoiding cone breakup. The mid and highfrequencies would be handled by a speaker having a smaller diaphragm less subject to breakup but unable to handle the large amplitudes associated with low frequencies. A high pass filter would be used to eliminate low frequencies from the small diaphragm speaker. The total sound spectrum can be broken up inthis way into two or more regions, with a loudspeakerand corresponding filterforeach region.

.- vLow frequency speakers can be constructed-that are .sub-

stantially free from cone breakup-upto aboutthree or four hundred cycles per second. Unfortunately, however, mid and high range loudspeakers employing cone diaphragms can not be made free from breakup effectsabove about or 600 cycles per second. Consequently other diaphragm designs have beendeveloped for the mid and high ranges.

One of the designs frequently employed for midrange speakersis the horn loaded annular diaphragm. With this'arrangement asinallring-shaped diaphragm closely coupled to the driving coil is coupled to a relatively large horn, generally of exponential flare. Breakup is minimized by the small diaphragm the horn loading being used to reduce the amplitude requirements. Horn speakers are subject to horn resonances, however, and horns are generally much larger and more expensive than equivalent direct radiator loudspeakers.

Another approach to the diaphragm breakup problem is the electrostatic loudspeaker. Here a diaphragm in the form of a thin film is suspended between perforated plates. The diaphragm is given a charge externally, and high alternating signal voltages are applied to the perforated plates. The charged diaphragm moves in response to the electrostatic forces and radiates sound through the perforations in the plates. The electrostatic speaker avoids breakup because the forces are applied uniformly over the entire area of the diaphragm and because the diaphragm is of uniform mass and is uniformly loaded by the air. There are several difficulties with electrostatic loudspeakers thathave prevented their application except for the extreme upper limit of the sound spectrum. Firstly, the electrostatic speaker is very directional owing to the large diaphragm necessitated by the small amplitude capability. Secondly, electrostatic speakers present a highly reactive load to the driving amplifier that reduces the effective efficiency to a very low value. Thirdly, the high charging voltage requires an undesirable high voltage power supply and introduces dust and moisture breakdown problems. Lastly, electrostatic speakers are relatively large, fragile and complex devices subject to maintenance difliculties.

Apreferred approach to the diaphragm breakup problem is the ribbon loudspeaker. FIG. 1 illustrates the elements of a prior art high frequency electrical-acoustical transducer in which a thin aluminum ribbon, l, is suspended longitudinally between the pole faces 2 and 3 of permanent magnet 4. Ribbon l is fitted closely between the pole faces so that the sound radiated from one side of the ribbon does not interfere significantly with the sound radiated from the opposite side. The ribbon is suspended by means of clamps 5 and 6 which support the ribbon at the ends thereof and supply intimate electrical connection therewith. Clamps 5 and 6 are connected to wires 7 and 7 respectively, which in turn communicate with the secondary winding 8 of the transformer 9. Typical dimensions for high frequency aluminum ribbons have been as follows: Thickness, 5 to 10 microns; width, one-half inch; length, 2 inches; electrical resistance, 10-20 milliohms. It is the purpose of the transformer 9 to match the relatively low resistance of the ribbon to the higher impedance of most amplifiers for sound reproduction which is typically 4-16 ohms. The primary winding 10 of transformer 9 is connected by means of

leads

11 and 12 to the output terminals of such an amplifier. The permanent magnet 4 should be designed to develop a relatively large field strength in the region of the ribbon 1, typical values lying between 3,000 and 10,000 gauss.

It can be seen that owing to the electrical continuity between the secondary winding 8 and leads 7 and 7, clamps 5 and 6 and the ribbon 1 that voltages induced in the secondary 8 by transformer action from the voltage applied to the primary 10, will cause a current flow in the ribbon l in a longitudinal direction. It is apparent that the interaction of the current through ribbon l and the orthogonally disposed magneticfield established by the magnet 4 will produce a force on the ribbon generally indicated by the numeral 13 and, in conformance with the laws of electromagnetic interaction, the force will be orthogonal to both the direction of current flow and to the direction of the magnetic field. Alternations in the current, corresponding to alternations in the original sound wave being reproduced, will produce corresponding alternations in the direction of the force as represented by the double-headed arrow 13. Since motion of the ribbon is unimpeded except by the surrounding air, its own inertia, and the end clamps, the ribbon will ideally move in response to the forces thereon and such motion will be communicated to the air and radiated as sound. It is the purpose of the transverse corrugations on the ribbon illustrated in FIG. 1 to provide a resilience that will allow the ribbon to be deflected alternately back and forth from its neutral position without being permanently stretched and without the imposition of excessive elastic restraint. Alternatively, transverse corrugations may be restricted to the terminal portions of the ribbon, the central portion being stiffened by longitudinal corrugations or dimpling. It is necessary that the clamps 5 and 6 provide intimate contact along the entire width of'the ribbon so that currents entering the ribbon by means of wires 7 and 7 will do so uniformly in order that all parts of the ribbon will be subjected to the same degree of current flow.

The uniform distribution of current within the ribbon, together with the uniform distribution of the magnetic field between the pole faces of the magnet ideally speaking, ensure that the force on the ribbon is likewise uniformly distributed throughout the entire area of the ribbon between the pole faces. Since the ribbon is of unifonn thickness, the mass of the ribbon is also uniformly distributed; and since the air is everywhere in contact with the ribbon, it can be seen that all restraints upon the ribbons motions are also distributed uniformly in a like manner to the force thereon, with the exception of the restraint imposed by each end clamp. As a result of the uniformity of distribution of force and restraint, the motion of the ribbon in response to the current induced therein is also uniform and the ribbon, although fragile and having little resistance to deformation, nevertheless moves substantially as a solid body executing in theory the vibratory motions imparted by the electromagnetic force in such a manner that every point of the ribbon moves in the same direction and to substantially the same extent as every other point.

Loudspeakers based upon the ribbon principle have been constructed from time to time and have displayed the freedom from diaphragm mode resonances that would be expected considering the uniformity of forces and loading upon the ribbon diaphragm. Unfortunately, standing waves induced by the restraints of the end clamps have seriously degraded the performance of such speakers below frequencies of approximately 2,000 cycles per second and rendered their use impractical at lower frequencies. Since low frequency speakers develop cone breakup above about 300 cycles per second there is considerable incentive to extend the range of ribbon type loudspeaker downward to less than 300 cycles per second, thereby to avoid resonance modes throughout the sound spectrum, and particularly in the speech range of 200 to 3,000 cycles per second.

l have found from experimenting with ribbon loudspeakers that uncontrolled longitudinal standing waves cause large spurious ribbon motions detrimental to the performance of the loudspeaker. At high frequencies where the motion of the ribbon may be only a few microns whereas the length of the ribbon may be several centimeters, it can be seen ribbon motions including standing waves are of small consequence. At lower frequencies however, the motionof the ribbon may become the ribbon imposed by the end clamps set up traveling waves within the ribbon and that the traveling waves are reflected by the clamps so that they interfere at discrete frequencies to fonn standing waves of much larger amplitude than the travel- I ing waves themselves. Such standing wave effects may occur at frequencies well below the nominal range of operation owing to practical limitations on the design of the described low pass filter, and may not cause noticeable irregularities in the frequency response of the ribbon loudspeaker. Nevertheless, large standing wave amplitudes may cause the ribbon to be displaced beyond the main body of uniform magnetic flux to the extent that forces on the .ribbon may no longer be proportional to the voice currents therein. Large standing wave amplitudes also produce objectional sound clatter owing to the free edges of the ribbon scraping-the magnetic poles. Deterioration of the ribbon from abrasion of the edges,

- stretching of the corrugations,- and fatigue at the end clamps has also been observed. The ribbon of a ribbon loudspeaker is relatively immune to damage from rough handling owing to the very low mass of the ribbon diaphragm. A rear cavity commuch greater and l have found that restraints on the motion of pression chamber may be acoustically coupled to the rear surface of the ribbon to prevent possible ribbon damage from wind, while the rear cavity in combination with a high pass filter may be employed to prevent possible ribbon damage from low frequency electrical impulse transients. Nevertheless, it is a matter of record that prior art ribbon type loudspeakers are considered to be fragile and subject to damage from low frequency transients. However, I have discovered that such damage is due entirely to the effects of standing waves induced in the ribbon by the described action of the end clamps.

It can be seen from the foregoing that it is very desirable that ribbon loudspeakers should be free to operate down to less than about 300 cycles per second without the production of destructive and performance degrading longitudinal standing wave patterns in the 'ribbon.'lt is an important object of my invention to reduce standing wave disturbances to the point where a ribbon loudspeaker may operate down to a frequency of about 200 cycles per second in an economic manner so that ribbon loudspeakers may be marketed widely for the first time.

SUMMARY OF THE INVENTION In accordance with the invention, significant attenuation of the aforesaid standing waves in a ribbon loudspeaker is accomplished in one embodiment by providing means, in the form of a viscous substance applied to the suspended terminal sections of the ribbon, for virtually eliminating the reflection of wave energy at the ribbon clamps. In another embodiment, standing waves are attenuated by virtually eliminating the production of traveling waves in the ribbon by progressively reducing ribbon motion toward the clamps by increasing the mass of suspended terminal portions of the ribbon relative to more centralized portions. In a third embodiment, the generation of traveling waves is reduced by the use of means for decreasing the magnetic flux density at suspended terminal portions of the ribbon relative to more centralized ribbon portions.

High frequency and midrange ribbon loudspeakers built by the inventor and incorporating the first and third embodiments in combination have produced excellent results over thousands of hours of operation providing, to the best of my knowledge, a clarity in the reproduction of speech and music in the mid and high frequency ranges not obtainable by any other types of loudspeakers.

Other objects, features, and advantages of my invention will become apparent from reading the following material taken in conjunction with the drawings in which:

FIG. 1 discloses a perspective view of a prior art ribbon loudspeaker;

FIG. 2 discloses a first embodiment;

FIG. 3 discloses a second embodiment of my invention;

FIG. 4 discloses a flux diagram associated therewith;

FIGS. 5 and 6 disclose additional embodiments.

DETAILED DESCRIPTIONOF THE INVENTION I have discovered several methods of reducing the motion of the ribbon at suspended terminal portions thereof, independently of the action of the clamps, thereby reducing the generation of traveling waves which may, by' subsequent reflection at the clamps, become standing waves as described. One such method is to increase the mass of the terminal sections of the ribbon by grading the thickness of the ribbon material before corrugating from a thinnest value at the center ribbon, or by coating the suspended terminal portions thereof. It is most desirable to increase the thickness gradually toward the ends of the ribbon in a manner similar to that shown in FIG. 2 since mechanical discontinuity is the cause of both traveling wave generation and energy reflection. Usually both terminal sections of the ribbon would be made thicker than thecentral section. In an alternative method, mass per unit length of the central portion of the ribbon could be decreased after corrugation by stretching the corrugations more in the central section than at the ends. I

As shown in FIG. 6, viscous semisolid material 26 such as commercially available loudspeaker cone edge dope may be applied to the terminal sections of the ribbon. Such viscous material operates to reduce traveling waves by virtue of the mass of the material added to the terminal sections as described hereinbefore. Additionally the viscous material dissipates wave energy, that would otherwise be reflected at the clamps, by virtue of viscous action caused by the flexing of the viscous material, such energy being converted to heat. A viscous or dissipative coating is therefore a preferred means of increasing the mass of the terminal sections of the ribbon. While the application of viscous material would preferably be operative at both ends of the ribbon, it might be possible to achieve effective results by the application of such material to only one terminal section of the ribbon.

In other words, the application of viscous material, which of course possesses mass, especially but not necessarily as indicated in FIG. 6, reduces the aforesaid mechanical discontinuity associated with the clamp, which causes the generation of standing waves in the first place, while concurrently the viscous action dissipates reflected energy that may nevertheless be present owing to other discontinuities that may be present. In the alternative a non viscous but dissipative material such as felt or other fibrous substance may be used to convert mechanical wave energy into heat by coacting with ribbon terminal portions.

The thickness of any applied layer should preferably be progressively tapered from a thickest application adjacent to the clamp, to essentially zero thickness adjacent to the central section of the ribbon. Not only does such tapering essentially nullify ribbon loading discontinuities, but stressing of the dissipative or viscous material is made more uniform so that all of the material may be substantially equally effective in absorbing reflected energy, thus dissipating maximum energy with a relatively small mass of viscous material. Alternatively, viscous or dissipative material may contact a terminal suspended ribbon portion over a relatively small area to provide effective wave energy absorption.

Another means of reducing ribbon motion toward the ends thereof is to reduce the strength of the magnetic field progressively along the length of the ribbon from a maximum value near the center of the ribbon to a minimum or zero value at the ends thereof. Since the motion of the ribbon is a function of the strength of the magnetic field, motion of the ribbon will ribbon will likewise be reduced toward the ends by this means. In order to reduce the force at the ends of the ribbon to zero by this means, it is preferable that the magnetic field strength drop to zero at the ends of the ribbon, that is at the point of clamping.

FIG. 3 illustrates a magnet and clamp structure that satisfies the foregoing requirements closely. A front view of the ribbon and magnet structure is shown with ribbon l4 suspended between the magnet poles l6 and 17. In FIG. 3 the sides of the magnet poles are shown as being extended by nonmagnetic sections I8 and 19. By virtue of the nonmagnetic material the flux density is reduced at the terminal sections of the ribbon to some extent, owing to the fringing of the magnetic flux. But flux density does not fall off very rapidly beyond the limits of the ends of the

magnet pole pieces

16 and 17. Howevenby constructing the ribbon support means or clamps of magnetic material such as iron or mild steel, the flux density is brought to zero at the ends of the suspended ribbon owing to the shunting action of the ferromagnetic clamps. FIG. 4 illustrates a desirable resultant profile of the flux density B as plotted along the ribbon axis. I have found that this arrangement most effectively attenuates longitudinal standing waves up to about 300 cycles per second in the case of a ribbon of about 6 inches in length. On the other hand the application of viscous material to the suspended terminal portions of such a ribbon most effectively attenuates higher frequency standing waves. Accordingly, by combining these approaches I have built inexpensive, true high fidelity ribbon loudspeakers, that operate down to 200 cps. To my knowledge this has not previously been accomplished.

The reduction in flux density toward the ends of the ribbon should not be appreciably more abrupt than shown in FIG. 4 since a sharp termination of the magnetic field produces a discontinuity in the force along the ribbon leading to the further generation of standing waves. The flux profile is advantageously shaped to approximate one half of a sine curve. While it is highly preferable that the flux drop to zero just at the ends of the suspended portion of the ribbon, some benefit may accrue by having the terminal portions of the suspended ribbon extend beyond the poles of the magnet without shunting so that the flux density is merely reduced at the clamp.

An alternative approach to the control of magnetic flux density along the ribbon is indicated in FIG. 5. In this embodiment a magnet of high coercive force and relatively low incremental permeability is employed, preferably in the form of two

pole pieces

16 and 17, attached to a soft iron frame or yoke that completes the magnetic circuit. Such magnets can be magnetized differentially along the length thereof, and the flux density in the central and terminal portions can be adjusted separately. By magnetizing the central portion more strongly than the terminal portions 21 flux distribution of the shape required for the reduction of standing waves may be produced. Magnetic shunts may also be used with this arrangement, and in fact the extension of the poles by nonmagnetic material may also be employed if desired. A flux density distribution along the length of the ribbon, approximating one half of a sine curve produces the mosteffective result.

Obviously, many modifications and variations of the present invention are possible in the light of the above teachings. It is therefore to be understood, that within the scope of the appended claims, the invention may be practiced otherwise than as specifically described.

I claim:

1. In a ribbon loudspeaker having a sound radiating conductor suspended between magnetic poles, said sound radiating conductor having supported terminal sections and a suspended centralized section therebetween, the improvement comprising: motion reduction means coextending longitudinally with said suspended centralized section from each terminal section to a point less than half way across said suspended centralized section, and coacting with said terminal sections for reducing the motion of said suspended centralized section to a degree which progressively decreases from said supported terminal sections toward the center of said suspended centralized section.

2. The combination as set forth in claim 1 wherein said motion reduction means comprises means for reducing the electromotive forces applied to end portions of said suspended centralized section of said sound radiating conductor relative to the electromotive forces applied to more centralized portions thereof.

3. The combination as set forth in claim 1 wherein said motion reduction means includes means for reducing the magnetic flux density in the region of end portions of said suspended centralized section of said sound radiating conductor relative to more centralized sections thereof.

4. The combination as set forth in claim 3 wherein said means for reducing the magnetic flux progressively decreases the intensity of said flux along the length'of said suspended centralized section of said sound radiating conductor to produce an absence of flux at the support terminal sections.

5. The combination as set forth in claim 3 wherein said means for reducing the magnetic flux comprises means for diverting flux produced by said magnetic poles away from the end portions of said suspended centralized section of said sound radiating conductor.

. 6. The combination as set forth in claim 5 wherein said means for diverting comprises at least one magnetic shunt positioned to virtually eliminate flux at the supported terminal sections of said sound radiating conductor.

7. The combination as set forth in claim 1 wherein said motion reduction means comprises means for increasing the mass per unit length of the end portions of said suspended centralized section of said sound radiating conductor relative to more centralized sections thereof.

8. The combination as set forth in claim 3 wherein said means for reducing the magnetic flux'comprises a pair of magnetic shunts and means for positioning said shunts beyond the suspended centralized section of said sound radiating conductor for sharply reducing the magnetic flux applied to end portions of said centralized section of said sound radiating conductor.

9. The combination as set forth in claim 8 wherein said magnetic shunts constitute support means for supporting said sound radiating conductor.

10. The combination as set forth in claim 1 wherein said motion reduction means comprises means for dissipating mechanical energy of wave motion of the end portions of said suspended centralized section of said sound radiating conductor.

11. The combination as set forth in claim 10 wherein said means for dissipating comprises a mass of viscous material positioned upon the end portions of said suspended centralized section of said sound radiating conductor.

12. The combination as set forth in claim 3 wherein said motion reduction means includes a mass of viscous material positioned upon the end portions of said suspended centralized section of said sound radiating conductor.

13. The combination as set forth in claim 1 wherein at least one pole is magnetized to produce reduced flux density at the end portions of said suspended centralized section of said second radiating conductor relative to the flux density at more centralized portions thereof.

14. The combination as set forth in claim 13 wherein said magnetized pole has a flux density distribution along the length thereof which approximates half of a sine wave.

15. The combination as set forth in claim 13 further including means for dissipating mechanical energy of wave motion of end portions of the suspended centralized section of said sound radiating conductor. p

16. The combination as set forth in claim 15 wherein said means for dissipating comprises a mass of viscous material positioned uponthe end portions of said suspended centralized section of said sound radiating conductor.

Claims

4. The combination as set forth in claim 3 wherein said means for reducing the magnetic flux progressively decreases the intensity of said flux along the length of said suspended centralized section of said sound radiating conductor to produce an absence of flux at the support terminal sections.

6. The combination as set forth in claim 5 wherein said means for diverting comprises at least one magnetic shunt positioned to virtually eliminate flux at the supported terminal sections of said sound radiating conductor.

8. The combination as set forth in claim 3 wherein said means for reducing the magnetic flux comprises a pair of magnetic shunts and means for positioning said shunts beyond the suspended centralized section of said sound radiating conductor for sharply reducing the magnetic flux applied to end portions of said centralized section of said sound radiating conductor.

15. The combination as set forth in claim 13 further including means for dissipating mechanical energy of wave motion of end portions of the suspended centralized section of said sound radiating conductor.

16. The combination as set forth in claim 15 wherein said means for dissipating comprises a mass of viscous material positioned upon the end portions of said suspended centralized section of said sound radiating conductor.