US1787055A - Acoustic diaphragm - Google Patents

Description

Dec. 30, 1930. v. A. SCHLENKER 1,787,055

ACOUSTIC DIAPHRAGM Filed March 16, 1928 Fig.4

vwentoz VESPER ANDERSON SCHLENKER 35% his, 61mm,

Patented Dec. 30, 1930 VEBPER mnnnson scnLENxnn, or omnen, unw JERSEY acous'nc nrernmeu Application filed larch '16, 1928. Serial No. 262,290.

This invention relates to sound reproducing apparatus and more particularly to a diaphragm for converting mechanical vibrations into sound, and has particular application in loud speakers utilized for converting electrical waves or oscillations into sound waves correspondin thereto.

In the past it has een customary to utilize some form of electrical vibrating device such '10 as an electro-magnetic or electro-dynamic motor for vibrating direct acting diaphragms. It has been realized that the sound waves produced by such arrangements are by no means faithful reproduction of the electrical waves supplied to the driving motor, since distortion is introduced to a greater or less extent. A portion of such distortions may occur in the motor element itself, particularly when it is of the so-called electromagnetlc type, in which a vibrating arma ture, sprin biased to one position, is utilized.

In addition, distortion also occurs in the diaphragm itself, and while the causes of distortion in the motor elements are, to a certain extent known, in my opinion, prior to the present invention, very little has been known about the distortion produced by the diaphragm itself.

It is an object of this invention .to provide a diaphragm which is in itself capable of giving an extremely faithful reproduction of'the vibrations impressed thereon: that is to say, a diaphragm which will deliver to the air an extremely faithful copy in the form of sound waves of the mechanical vibrations impressed upon it.

It is a further object of this invention to provide an acoustic diaphragm in which the distortions heretofore present in diaphragms are eliminated or substantially reduced.

It is a further object of this invention to provide a self-supporting metallic diaphragm of suflicient size to operate directly upon free air, having a mass which is negligible or small in comparison with air loading thereon at sound frequencies less than 1000 cycles.

It is a further object of this invention to provide a self-supporting diaphragm, substantially devoid of inherent damping and inherent stiffness and structurally stiffened by being formed in a manner which distributes the structural stiffness throughout the diaphragm in a particular way, to be here after described.

It is a further object of this invention to provide a self-supporting direct-acting diaphragm having very low inherent damping and inherent stiffness, and having a propagation constant which is substantially independent of frequency and approaches the propagation constant of free air.

Still other objects and advantages of my invention will. be apparent from the specification.

The features of novelty which I believe to be characteristic of my invention are set forth with particularity in the appended claims. My invent-ion itself, however, both as to its fundamental principles and as to its particular applications will best be understood by reference to the specification and accompanying drawing, in which Figure 1 is a perspective view of a surface having the form of a hyperbolic paraboloid.

Figure 2 isa sectional view of a diaphragm generated by the revolution of a curve having the equation Y=Log mac.

Figure 3 is a graph showing the generating curve: and,

Figure 4 is a side elevation partly in section showing one form of a loud speaker according to my invention.

In the usual form of direct-acting or socalled sounding-board loud speakers, it is usual to impress the mechanical vvibrations which are to be converted into sound, upon the diaphragm or vibrating surface at substantially a single point, or at any rate. at an extremely small part of the area of the diaphragm. The mechanical vibrations so impressed, travel to all parts of the diaphragm and the vibrations of the various elements of the diaphragm set into vibration the air adjacent thereto. I

It is known that extremely complex vibrations, corresponding to speech or music may be resolved into a number of sinusoidal vibrations of different frequencies and amplitudes, the resultant of which correspond the particular speech or music wave. The diaphra gm may thus be regarded with respect to, the transmission of vibrations from the point of application to other parts thereof, as a mechanical power transmission line, which is subjected to sinusoidal vibrations of various frequencies and amplitudes, and which vibraogous case of a telephone transmission line.

As is understood, the telephone line itself has definite characteristics, such as inductance, capacity, resistance, leakage, and attenuation, the attenuation being a function of the other constants of the line, and also of frequency.

'In the case of a telephone line, it is naturally desired to keep the leakage as low as possible; in'other words, to deliver as high a percentage of the input power at the output end as is possible. In this respect, the functioning of an acoustic diaphragm does not correspond With a power transmission line, since in an acoustic diaphragm, it is desired to make the radiation, which may be said to correspond to the leakage of a telephone line, as high as possible.

Considering the distortion which is introduced after the vibrations are impressed upon the diaphragm, it may be stated that in general they fall within two classes: (1) distor tion which may arise within the diaphragm itself, due to various causes, which will be discussed later: and (2) distortion and loss of efficiency which may take place in the radiation from. the diaphragm to the atmosphere, in a manner which will hereinafter be described more in detail. The first of these effects, namely the distortion within the diaphragm itself, arises because the fiexural waves set up in the material of the diaphragm travel at a speed which, to a certain extent, varies with the frequency of the impressed wave, and because the attenuation varies with the frequency; such distortions are, in general, analogous to the distortion occurring in the transmissions of telephone currents along a telephone line, wherein the waves travel at a speed which is equal to Where p represents frequency, and ,8 is a function of the constants of the line, and the frequency, called the wave length constant.

The amplitude of currents travelling along such a line is decreased in the ratio of 1 e' per mile of transmission line, where a is rents travel along such a line, thecurrents of various frequency are dephased to a greater or less extent, depending upon the constants and length of the line and frequency, and are .also changed in their relative amplitudes, the

higher frequencies being attenuated to a greater extent than the lower frequencies. This distortion, if permitted, will ultimately render speechcompletely unintelligible.

Unless prevented by means suchas I contemplate, a similar elfectwill take place as the complex vibrations impressed at one point on an acoustic diaphragm travel to the other points of the diaphragm: in the case where the velocity of propagation of the fiexural Vibrations is different at different frequencies, phase distortion will occur, as will be readily understood, and amplitude distortion will occur in case the inherent damping or energy loss in the diaphragm is different for diiferent frequencies, so that if a complex vibration beimpressed at one point on the diaphragm, the vibration of another point at a distance from the first point,

will not be truly similar to the vibration of the first point. Under such conditions, naturally the vibrations impressed upon the air will differ at different points of the diaphragm, and the difference or distortion will, in general, be greater the larger the size of the diaphragm, and the greater the difference in velocity of wave propagation and attenuation with frequency. This distortion is particularly noticeable in the-case of diaphragms of non-metallic material, such as paper cones and fabric diaphragms, and becomes increasingly objectionable as the size of the diaphragm is increased. It is evidenced by a mufiied tone, oor articulation, suppression of the higher requencies and a general lack of clarity and tonal definition, giving speech an unnatural quality.

In order'to minimize this distortion, it is advisable to arrange the diaphragm in such a way thatv waves of all frequencies travel at the same speed in the material of the diaphragm, and further, that all frequencies attenuate in the same ratio. However, while such an arrangement would eliminate the distortion within the diaphragm itself and in its radiation, itwould not necessarily eliminate a part of the second undesired efi'ect, namely, loss-of efficiency in radiation from the diaphragm. his second. effect may be characterized asdue, in general, to the speed of travel of flexural waves in the diaphragm being different from the speed of travel of sound in the air. .An exthe diaphragm" occurs at that point and a speed of soundin air. Now,.it willbe apparent that ifthe flexural waves in'thediaphragm travel at the samespeed, the original sound wave in air will be continuously absorbing energy from the'diaphragm at the proper point 'on'the wave front: however,

material is greater or less than thespeed of sound in air, the sound wave in an running across the face of the'diaphragmwill be displaced with respect to the corresponding fiexural wave travelling in the material of the diaphragm itself, and energy will be supplied from the diaphragm to the air with an improper phase relation, which will cause reinforcement of the sound waves in the air at certain points on'the diaphragm and partial neutralization at otherflpoints.

Theresultof' this is that as the ratio of Ve--' Moreover, if the velocity of propagation'of.

flexural waves in the diaphragm varies with the frequency, the radiation will be different for different frequencies, being least eflicient on frequencies at which the velocity: of flex ural variations difl'ers most widely 0111 that of the velocity of sound in air. This, will cause partial suppression of such frequencies. Folthese reasons, it is desirable not only to arrange the diaphragm so that flexural waves of all frequencies travel at the same speed and attenuation at the same rate,

but also so that the velocity of propagation of flexuralwaves in the material should equal as closely as possible that of sound waves in If this is done, the distortions occurair. ring after the vibrations are impressed upon the diaphragm are eliminated or minimized, and ifthe diaphragm be operated by a driv ing motor which is substantially of disunit of line length. Under this condition,

-tortion, an extremely pure reproduction of sound will he obtained.

The general condition for distortionless transmission in a telephone line is UR LS, where C represents capacity: R, resistance; L, inductance :-and Sl, leakage per currents of all frequency travel at the same speed and attenuate at the same rate. In generaL. the same is. true for an, acoustic diaphragm, but in this instance, the proper physical constants must be used: capacity corresponds to stiffness: resistance to inheradapted to the purpose. t

The diaphragm, according to my invention, the

entdamping of the material itself, such as occurs if the diaphragm be operated in a vacuum inductance to mass: and leakage to radiation. 1

Y It is essential that the diaphragm be critically damped in order to preventany tendency to resonance of the diaphragm as a whole or 'in parts, and' in order to prevent transient vibrations. This critical damping is obtained by utilizing material having minimum surface density, and by arranging the diaphragm so that it has a high radiation resistance by :virtueof close coupling to the air.

if the speed of travel of flexural waves in the qulred, the material'will be so thin as to have Because of minimumsurface density revery low inherent stiffness, and in order to ob tain high radiation and a velocity of wave propagation approaching that of sound. in

air,"sti'tfness must be imparted to the diaphragm by structural formation. W'hile a high degree of stiffness can be obtained by increasing the thickness of the diaphragm,

and whilefthis the normal way of increasing the stiffness of'a diaphragm, it is undesirable in the present instance, and this is accomplished by forming the diaphragm ina special shape, which I find to be is preferably formed of a metal havin least possible mass, such as aluminum a loys,

and is formed from sheets so thin as to have very low inherent stiffness. The material is 'ven what I term-structural stiffness by orming the diaphragm in such-a manner that it as reversed curvature at every oint its periphery: that is to. say,i two planes at right angles to each other be passed through the diaphragm,'the intersection between th'e diaphragm and one planewill be curved continuously in one direction, and the intersection between the other plane and-the diaphragm will be curved continuously in the opposite-direction; This particular curvatum is given-the diaphragm so that for any deformation normal to the diaphragm at any point, compression will be set up along the line of one curvature, and tension will be set along the line ofithe other. curvature,'and

if the dlrection of deformation be reversed,

the compression and tension components will be reversed, that curvature which was formerly under tension being under compression and vice versa; the result of this is the formation of a dynamically stable surface which particularly An example of such a surface isthe hvperbolic paraboloid, or saddle shaped surface, well known to mathematicians. I believe that the blasting, which may take place at any one form to another. For example, consider a small disc of metal which is dished in one direction so as to have a slight curvature. It will be understood-that there are present certain strains in the material which tend to make it hold such a dished shape. If now, the disc be pressed in such a manner as to cause it to dish in the opposite direction,- it

' will resist deformation up to a certain point.

paper cones.

and then will suddenly swing over to the other condition in which the dishing is reversed,

producing a sudden snap. .An analogous condition I believe to exist in the well known Any elementary area of such a cone is dynamically unsymmetrical and un-' stable, since the forces which come into play for an inward deformation at any point on the cone are diflerent from those which come into play for an outward deformation thereof. If a stress resulting from deformation be resolved into radial and circular components, it will be seen that for an inward deformation at-any point on a cone, tension is produced radially and compression circumferentially, whereas for an outward deformation of the same point, tension is still produced radially and tension is also produced circumfer I prefer to form the diaphragm according" to my invention of material having the least possible mass and damping within itself, and

for this reason I utilize metal, preferably aluminum alloys, having minimum mass and relatively high strength. In order to maintain the mass at the lowest possible amount,

and thereby permit 'maximum-radiation, In

utilize metal of thegreatest possible thinness,

, preferably not more than a few thousandths thick. Satisfactory results have been obtained by the use of such alloy 0.002 thick and having a mass not exceeding 0.03 grams per square centimeter. From a purely theoretical standpoint, it might be desirable under certain conditions to use even thinner material but practical difliculties of working extremely thin material will usually be too great to obtain uniformly satisfactory results. Y

The diaphragm may be conveniently made in the formof a surface of revolution, such as the logarithmic surface of Figure 2; In this instance, the surface may be considered as be-' the rotation of a curve rep-.

ing generated b resented by Y =Log mm, about an axis R--R, which does not intersect the curve within the generating segment. The surface so produced will thus be seen to have the domed portion, or in other words the angle a between the inner ed e of the diaphragm and the m. axis (Figure 2 is approximately 50,

whereas the angle 7 between the outer por tion or edge of the diaphragm and the :v axis,

is aboutl0 to 15. The constants are likewise so chosen that the maximum curvature in the radial direction occurs adjacent the outeredge of the diaphragm, as at point C.

It will be noted that under these conditions, the radial curvature is greatest where the circumferential curvature is smallest and vice versa. In other words, if the radial curvature be deslgnatedas C, and the circumferential curvature or curvature normal 7 to the radial curvature, be designated as O then it will, in general be true that It should be pointed out that it is not necessary that this condition be fulfilled exactly, but, satisfactory results are obtained when it is approached. For example, the curvathousand to one would be quite noticeable. j

I Other curves in addition to those already described, may be utilized in the formation of the diaphragm. For example, the radial curvature may be in the form of a cycloid or aparabola, but in any case if the surface is one" of revolution, it is desirable to have the greatest radial curvature adjacent the outer periphery of the diaphragm, since at that point) the smallest.

Referring now more particularly to Figure;1,"I have shown a surface in the form of circumferential curvature is a hyperbolic para-baloid whichis a so-called ruled surface rather than a surface of revolution, from which it will be evident that any elemental area within the periphery of the diaphragm is reversely curved, the curvature along one plane being convex, and the curvature along another plane at right angles thereto being concave.

Referring now more articularly. to Figure 2, in which I have s own avsection of a diaphragm comprisinga surface of revolution: OY desi hates the axis of the diaphragm and O the axis normal thereto. The constants and segment of the curve and the axis of revolution R-R are so osen and the size'of the diaphragm is so chosen diaphragm.

with respect to the constants of the curve,

that the, angle a is approximately 50, whereas the angle y-is from to The limits of the generating curve are so placed'that the p'oint'of maximum curvature, such as C, lies adjacent the'outer periphery p of the The driving force is preferably applied around the circle generated by the pointD,

as the'generating curve is revolved about the axis R-R.

Referring now more particularly to Figurei 3, I have shown the generating curve whose equation is Y=Log mm, the curve followin the points 1, 2 and 3. It will be noted that t e curve 'crosses the 2: axis at :1 point where a.- isequal. to unity.

Referring l now more particularly to Figure 4, 5 designates a diaphragm such as shown in Fi re 2, and 6 is a hollow frustroconicalmem r suitably attached to the apex of the diaphragm, as by cementing, and 7 is ,a driving rod secured to the hollow'ifrustro- 'coni'cal member 6,. as by nuts (not shown) or in any other sluitablemanner: 8 designates a driving motor of any suitable type, such for 7 example as a balanced armature motor. It will be understood, however, that any other example, the so-called type of driving motor may be utilized, as for dynamic type of drive. The form of drive forms per se .no

part of the invention claimed herein, and is therefore notshown indetail.

'While the actual size of the diaphragm is not of great importance, it will be understood that the distortions, which it is the object of this invention to overcome, are more I noticeable in the prior art diaphragms' as the size is increased. If the diaphragm accord- "ing to my invention is of sufiicient size so as notto render desirable the use of a bafile vplate, it-may be operated in free air sup- 'portedia's its center and with its periphery entirely free.' If, however, the size of the diaphragm is decreased to such'an extentthat there is an appreciable loss of low fre quencies when the diaphragm is so 0 erated,

it is preferable to provide some 'baflle arrangement, such as that shown in the orm of drawing wherein '9 designates diagrammatically asuitable form of baflle provided with an opening therein and having its inner edge 10 reversely bent. For the purpose of pro- .viding an air seal betweenlthe bafile' plate 9 (and the diaphragm, a ring or annulus of '-flexible'material 11, such as plush, linen,

leather or'the like, maybe cemented to the periphery of the diaphragm -and secured against the flange 10 by suitable tension members, such as several turns of string. 12.

v particular construction forms per se no part of my invention.

. While Ihave shown and described certain preferred embodimentsof my invention, 1t is thereto.

understood that modifications and changes may be made without departing from the spirit and scope of-my invention, as will be understood by those skilled in the art, and

also while I have described the use of my invention in a loud speaker for converting mechanical vibrations into sound, it is apparent that it may equally well be utilized as amicrophone diaphragm for the purpose of converting sound waves into mechanical motion, or electrical currents corresponding I claim a v v 1. A soundreproducer comprising a self supporting, direct acting diaphragm of thin metal, of such shape that any, section containing the'axis is a curve convex .toward the said axis and any section made by a plane normal to the axis is a curve concave towardthe axis,

the mass and area of said diaphragm being i such that the fundamental period thereof when vibrating freely as a bell is below the lowest important voice frequency.

2. A sound reproducer comprising a selfsupporting, direct acting diaphragm of relatively thin metal in the formof a surface so formed that-deformation of anypoint within the periphery of said diaphragm in a direction normal to the surface at said point produces tension in one componentand compression in a component normal thereto, and deformation of said pointinthe opposite di rection produces reversed stresses.

3. A sound reproducer comprising a selfsupporting, direct acting metallic diathereon at sound frequencies up to 1000' cycles. said diaphragm having relatively small inherent stillness and relatively high struc- "tural stiffness, and being so formed that the curvatures at right angles are opposite in sign.

5. A sound reproducer comprising a selfsupporting, direct acting metallic-diaphragm having a mass which is small in comparison with air loading thereon at sound frequencies up to .1000 cycles, and having relatively high structural stiffness and having a negative curvature in one component and a positive curvature in the component at right angles.

thereto, the curvatures being such that the product of curvatures at any point on said diaphragm is substantially constant.

6. A sound reproducer comprising a self supporting, direct acting metallic diaphragm iia rah

having a mass which is small in-coinparison with air loadin thereon at sound frequencies up to 1000fcyc es, and having low inherent stifl'ness and high structural stiffnessf n the form of a surface of revolution convexjto ward the axis'and having its greatest'radial tion of the curve generatin the outer 'peripheral portionof the diap ragmhaving a less radius of curvature than the portion of i the curve generating the p phragm toward the'center thereof.

New York and State of New day of March, A. D. 1928.

any point, and having an attenuation constant p so thin as to have very low inherent stiffwhich is substantially independent of frel ncy 8.- Aself-supporting, direct acting metallic diaphragm, the mass of which is small in i comparison with'air loading thereon at sound frequencies less than 1000 cycles, having low inherent stifi'ness and reversed curvature at any point,and having a velocity of propagation of flexural waves which is substantially independent of frequenc 9. A self-supporting, irect acting metallic diaphragm the a mass of which is small in comparison with air loading thereon at sound frequencies less "than 1000 cycles, having low inherent stiffness and reversed curvature at v any point, having an attenuation constant which is substantially independent of he quency, and having a velocity of propaga-.

tion of flexural waves which is substantially independent offrequency. a

10. The method of minimizing distortion in a direct acting acoustic diaphragm which comprises forming said diaphragm of metal ness, and stiffening said diaphragm by form ing the material thereof into aasurface having reversed curvature in directions at right angles to each other.-

1 1. The method of minimizing distortion in a direct acting acoustic diaphragm, which comprises forming said diaphragm of metal. so thin as to have very low inherent stiffness, and stiffening said diaphragm by forming the material thereof into a surface having reversed curvature in directions at right angles to each other in such manner that the product of curvatures at every point within the periphery of said diaphragm is substantially constant.

12.. An acoustic diaphragm having the shape of a surface ofrevolution formed by revolution of a line having a continuously changing radius of curvature, the portion of the line having the least radius of curvature generating the portion of the diaphragm surface adjacent the outer periphery thereof.

13. An acoustic diaphragm having the shape of a surface of revolution formed by revolution'of a logarithmic curve.

revolution of a logarithmic 14. An acoustic diaphragm having the shape of a surface of revolution formed by curve, the porortion of the dia- 7o Si ed at New vYork c1t 1n the count of. gm y, York, this tgnth 'VESPER ANDERSON S HLENKERQ y I v I I I

Images