US3393766A

US3393766A - Speaker system

Info

Publication number: US3393766A
Application number: US551028A
Authority: US
Inventors: Lawrence H Mitchell
Original assignee: American District Telegraph Co
Current assignee: American District Telegraph Co
Priority date: 1966-05-18
Filing date: 1966-05-18
Publication date: 1968-07-23
Anticipated expiration: 1985-07-23

Description

L. H. MITCHELL SPEAKER SYSTEM July 23, 1968 5 Sheets-Sheet l Filed May 18, 1965 L. H. MITCHELL.

SPEAKER SYSTEM July 23, 1968 3 Sheets-Sheet 2 Filed May 18, 1965 4o FIGA.

AMPLIFIER SPEAKER SYSTEM 3 Sheets-Sheet 3 Filed May 18, 1966 .mC zouwm mun. 335V Al 55:35

zw xm lOm United States Patent O 3,393,766 SPEAKER SYSTEM Lawrance H. Mitchell, Queens, N.Y., assignor to American District Telegraph Company, Jersey City, NJ., a corporation of New Jersey Filed May 18, 1966, Ser. No. 551,028 Claims. (Cl. 181-31) ABSTRACT 0F THE DISCLOSURE This invention relates to sound reproduction systems, and in particular to sound reproduction systems in which a suifuse or diffuse field of sound is produced over a relatively large area and in which a stereophonic effect is produced from a monophonic source.

In the design of sound reproduction systems, one of the most challenging problems is to deliver a substantial amount of sound power in a manner that will produce for a listener an acoustic sensation that resembles as closely as possible the acoustic sensation that the original sound would have produced. As pointed out by E. E. David, Jr., in The Reproduction of Sound, vol. 204, Scientific American, page 72 (1961), live sound is characterized by both auditory perspective and ambience, auditory perspective referring to the ability of the human hearing mechanism to locate a particular sound source, and ambience referring to the arrival at the listeners ears of many successive echoes of a sound from a number of different directions. Sound reproduction systems that attempt to provide one or both of these qualities in the reproduction of sound are generally referred to as stereophonic systems, and several of these systems are described in the above-mentioned E. E. David, Jr., article.

In stereophonic systems that provide both auditory perspective and ambience in the reproduced sound, ltwo separate sound channels are typically provided, each having its own microphone and loud-speaker. Two microphones are spatially separated to detect the original sound at two different points in the original sound field, and the two loud-speakers are also spatially separated to reproduce the original sound at two different points in the listeners environment.

Where the two microphones are placed close to the source of original sound, for example near the orchestra in a concert hall, the reproduced sound in the listeners environment may have excessive auditory perspective in that the reproduced sound impression is that which the listener would hear if he were located on the stage rather than in the audience. Since musical performances are generally directed at and for the benefit of an audience not necessarily located on the stage and further to take advantage of the acoustical characteristics of the concert hall, the degree of auditory perspective in many stereophonic recordings is found objectionable by some listenets.

In so-called quasi stereo systems, the illusion of stereophonic sound is created by producing only the quality of ambience in the reproduced sound, which is found by many listeners to be preferable to the stereophonic 3,393,766 Patented July 23, 1968 reprod-uction previously mentioned. This preference may be `attributable to the close approximation of the quasi stereo sound to the sound that is heard by the audience in a concert hall, where reverberations and multiple propagation paths for sound waves provide a full an-d live sound without distinct auditory perspective, that is, without location of a lparticular in-struments sound with respect to its position on the stage.

Ambience alone can be obtained in a quasi stereo system from a single or monophonic source, 'for example, a single microphone or recording channel, by supplying each of two or more `spatially separated loud-speakers with a different version of the same sound, for example, different artificial echo patterns of the same sound. By this arrangement, there is created for the listener a sound field in which different components of the sound arrive at the listeners ears from different directions or at different times or both, thereby affording an ambient quality that is characteristic of live sound. However, the production of different sound patterns from a monophonic source typically requires complex, expensive equipment.

In the present invention there is provided a sound reproduction system in which two sets of sound patterns are reproduced from a single sound source by an inexpensive speaker system of unique but relatively simple structure, thereby to create a diffuse sound field which provides the illusion of stereophonic sound. In the speaker system of this invention, a single electroacoustic transducer is positioned within a hollow tube or duct open at each end so that the interior of the tube is divided into two separate air columns, one column on one .side of the transducer and the other column on the other si-de of the transducer. By positioning the transducer so that the column of air on one side of th-e transducer is approximately twice the length of the air column on the other side of the transducer, and so that each motion of the transducer simultaneously produces a compression Wave or pulse of air in one column and a rarefaction wave or pulse of air in the other column, a two-to-one relationship is established between the fundamental resonant frequencies of the two sections of the tube, with the fundamental resonant frequency of the longer section being half the fundamental resonant frequency of the shorter section. It was discovered that a pronounced stereophonic eff-ect is obtained with this arrangement, because when the two air columns are excited by vibration of the transducer, the two air columns resonate individually and in combination at fundamental and harmonic frequencies. As a result, in the sounds emanating from the open ends of the tube, different frequencies predominate at each end of the tube.

An important feature of the speaker system of the present invention is its smooth and extended low frequency response, since the relatively long coupling path of air between opposite sides of the transducer minimizes cancellation of sound waves produced on one side of the transducer by simultaneously produced sound waves on the other side of the transducer.

Further, it has been found that the efiiciency of the transducer in the speaker system of this invention, in terms of converting electrical energy into sound energy, is substantially lgreater than that which is obtainable with the same transducer mounted in a conventional enclosure, since smooth frequency response is obtained in the present invention without resorting to back-loading or otherwise damping the excursions of the transducer as in conventional enclosures.

Those skilled in the art will immediately recognize from the following description and appended drawings that the principles of the present invention may be applied with advantage to various types of sound reproduction and transmission systems, it being understood that the description is intended for illustrative purposes only and in no manner to limit the invention.

The invention will now be described in greater detail with reference to the appended drawings, in which,

FIG. l is an elevation view, partially in section, of the speaker system of the present invention;

FIG. 2 is an elevation view of the speaker system of the present invention illustrating one mode of flushmounting the speaker;

FlGS. 3A, 3B, 3C and 3D are perspective views of speakers of various cross sections embodying the principles of the present invention;

FIG. 4 is a block diagram of a complete sound reproduction system embodying the principles of this invention; and

FIG. 5 is a -graph of assistance in explaining the present invention.

Referring now to the drawings wherein like reference characters designate like parts throughout the several views, there is shown in FIG. l a speaker embodying the principles of this invention, in which an electroacoustic transducer is positioned within a hollow tube 11 by means of annular mounting sections 12a and 12b to divide the interior of tube 11 into two separate air columns, respectively denoted A and B. Transducer It) may be of the electro-dynamic, direct radiator type havingl a paper diaphragm or cone 100 and a moving coil driving mechanism 101 to which a single or monophonic incoming electrical signal from a sound source is applied by way of input lead 152 and terminal 103. However, it will be understood by those skilled in the art that other types of transducers may be employed in accordance with the principles of this invention to provide the stereophonic effect described herein.

In order to obtain the desired stereophonic effect, transducer 10 must be positioned within tube 11 so that the length of the air column A on one side of transducer lfb is approximately twice as long as the air column B on the other side of transducer I0. This two-to-one relationship is indicated in the several views by denoting the length of the longer air column by the symbol X, and by denoting the length of the shorter air column by the corresponding symbol X/Z. Although tube 11 is shown as a unitary hollow member whose interior is divided by transducer 10 into two separate air columns, it is to be understood that the same result may be achieved by coupling together two separate hollow members, one on each side of transducer 10, where one of the hollow members is twice as long as the other. For example, tube 11 may comprise two separate tubes joined together along line 18, in which case the tube containing air column A on one side of transducer l0 is substantially twice as long as the tube containing air column B on the other side of transducer 10.

Because of the two-to-one relationship between the lengths of the air colu-mns A and B, the fundamental resonant frequency of the longer air column, A, is onehalf the fundamental resonant frequency of the shorter air column, B. Since the transducer 10 separating the two air columns is not an immovable barrier but is instead an active driving element producing simultaneous compressional pulses and rarefactional pulses in the air col- -umns on opposite sides of the transducer, pressure nodes and pressure peaks shift position within the two air columns resonating individually and in combination as the frequency of energy driving the transducer is changed. As a result, sound of a particular frequency is predominantly radiated from one end of tube 1I. The frequency selective characteristic of the speaker system of this invention may be demonstrated by connecting a variablefrequency tone lgenerator to the speaker, in which case sound will predominantly emanate first from one end, then from both ends, and then from the other end as the audio spectrum is swept. This frequency selective characteristic is `most apparent at lower frequencies where the harmonic mode of resonance is of low order. FIG. 5 illustrates graphically the frequency selective characteristic of the speaker system of this invention. The `graphs shown in FIG. 5 were obtained from measurements made by placing a microphone at each open end of a tubular speaker of the type shown in FIG. l, the speaker having a length of four feet and an inside diameter of two and one-half inches. Each microphone was spaced approximately one-quarter of an inch from the corresponding open end of the tube, and to each of the two microphones there was connected an amplifier and an alternating current voltmeter. To the transducer I0 inside the speaker, there was connected an adjustable frequency oscillator and the transducer was excited at each of a number of frequencies. The corresponding sound pressure level at each frequency, represented by the voltage of the output signal of each microphone at the open end of the tube, is tabulated in Table I below.

TABLE I A(2X End of Speaker) B(X End of Speaker) (Volts) (Volts) Frequency (c.p.s.):

1 Not plotted in FIG. 5.

The graphs shown in FIG. 5 represent the sound pressure levels in Table I plotted as a function of frequency. FIG. 5 makes graphically evident the manner in which the frequency selective characteristic of the speaker of this invention produces a stereophonic effect through ambience. Thus, there are large numbers of frequencies at which a pressure peak or maximum at one end of the speaker coincides with a pressure null or minimum at the other end of the speaker; pressure maxima or minima are not limited to one end of the speaker or the other, but appear at both ends of the speaker; and maxima and minima tend to alternate with one another along the frequency scale at each end of the speaker. It is to be noted in FIG. 5 that peaks and nulls do not always coincide exactly, which is believed to be attributable to both slight mispositioning of the transducer within the tube, and a finite but indeterminate acoustical width of the transducer, especially during excursions of the diaphragm in response to external excitation, both of which indicate that an appreciable deviation from the desired two-to-one ratio will degrade the stereophonic effect.

Since both human speech sounds and musical sounds are complex combinations of large numbers of frequencies, the frequency selective characteristic of the speaker of this invention causes certain portions of such sounds to be radiated predominantly from one end of the speaker o1- the other, so that different components of a sound arrive at a listeners ears from different directions over different propagation paths. As a result, sound from a monophonic source is reproduced by the speaker of this invention with the ambient quality characteristic of stereophonic sound reproduction systems.

In order for reproduced sound to be a faithful replica of the original sound, each component must be reproduced without si-gnicant alteration in intensity, and without the introduction of any degrading, spurious components in the reproduced sound. While the use of resonant air columns in this invention might lead to the immediate assumption that its frequency response would lack uniformity, it has been `found that the frequency response is smooth and distortion-free when the transducer is positioned in the manner shown in FIG. l and described above, to divide the interior of the tube into two separate air columns whose lengths are in two-toone relationship. At the same time, it has been observed that a substantially greater amount of sound energy is obtained from the same amount of input signal energy than is obtainable with the same transducer mounted in a conventional enclosure such as an infinite batiie enclosure. Although this phenomenon is not fully understood` it is believed that the increased sound energy is at least partly attributable to use of output energy from both sides of the diaphragm of transducer 10 in this invention; whereas in an infinite batiie the energy from one side of the transducer is deliberately trapped and prevented from entering the sound tield, thereby to preclude cancellation of the front wave of the transducer by the back wave of the transducer at lower frequencies, and also to load the diaphragm of the transducer to reduce unwanted resonance of the diaphragm, which results in compression of the excursions of the diaphragm.

At the ends of tube 11 there are provided

screens

13a, 13b to protect transducer 10 from unwanted foreign objects.

Screens

13a, 13b may be titted over corresponding telescoping sections 14a, 14b and then secured, for example, with cement, to the interior of tube 11.

The parameters of tube inner diameter, tube length, tube material, and speaker size may be varied within rather broad ranges, while retaining the two-to-one relationship between the air columns within the tube to achieve the desired stereophonic effect, increased transducer efliciency, and uniform response over a relatively broad frequency range. For tube irmer diameter, using a tube of circular cross-section as shown in FIG. 3A, it has been found that both 2% and 8" diameters, together with transducers of corresponding diameters, are satisfactory. With an inner diameter of 2%, together with a transducer having a cone 21/2 in diameter mounted in a basket (not shown) which is 2%" in diameter, it has been observed that this combination, when secured to a ceiling ten feet above floor level with the tube axis 112 parallel to the tioor provides on the order of four times as much sound coverage for a given gradient of sound level as an 8 diameter speaker mounted in a cotnventional enclosure at the same height, in addition to providing a stereophonic effect unobtainable from a single speaker mounted in a conventional enclosure. This increased coverage is partially due to excitation of a large volume of air at a relatively low sound pressure level rather than the smaller, directly radiated cone of air typically excited at a higher sound pressure level by a transducer of the same size mounted in a conventional enclosure. Of course, it is to be understood that other tube inner diameters, in combination with corresponding transducer diameters, may be employed as desired.

Frequency response is a function of tube length, for a given transducer, with the upper limit of tube length being determined primarily by the reproduction of high frequencies, and the lower limit being detenmined primarily by the reproduction of low frequencies. At the high frequency end of the audible spectrum, frequency response falls otf quite rapidly with overall tube lengths greater than about six feet. For reproduction of low frequencies, it has been found that frequencies lower than one-half the unloaded resonant frequency of the transducer cone are uniformly reproduced by using an overall tube length on the order of four feet. It to be understood, however, that this lower limit on tube length is not critical, since uniform low frequency response has been obtained with an overall tube length as small as four feet for both 21/2" and 8 diameter transducer cones,

even though a shorter overall tube length on the order of three feet is also satisfactory for an 8" transducer cone.

'Ihe tube may be constructed of any one of a number of materials, the choice of material depending upon the particular color or timbre, if any, that it is desired to impart to the reproduced sound. The relationship of material to timbre is much the same as that found in the construction of organ pipes. For sound systems where it is desirable to reproduce sound without introducing any additional or modied tonal qualities, it has been found that a combination of chipboard and kraftpaper is a preferred tube material since it is relatively inexpensive and produces no distinct tonal qualities in and of itself.

Turning now to FIG. 2, this drawing illustrates an arrangement for Hush-mounting the speaker of this invention. Within a concealed space 27 adjacent to the space 28 where the reproduced sound is to be introduced, there is mounted tube l1, which is provided with a transducer It) positioned to divide the interior of tube into two separate air columns whose lengths are in the two-to-one relationship desc-ribed above. In order to introduce reproduced sound into the space 2S, hollow ducts 21a and 2lb extend tube 11 to screened ports 22a and 22h in the wall of the concealed space 27. Although FIG. 2 shows curved ducts which direct the reproduced sound through the same side of concealed space 27, it is to be understood that the reproduced sound may be directed through ports located in any desired portion of concealed space 27. It is important to note that the lengths of ducts 21a and 2lb must be taken into account in positioning the transducer l0 within tube 11 to obtain the desired two-to-one relationship between the air columns on either side of transducer 10.

Referring next to FIGS. 3A through 3D, these drawings illustrate that tubes of either circular or non-circular cross-sections may be employed in practicing this invention. FIG. 3A shows a tube 11 of circular cross-section, FIG. 3B shows a tube 300 of rectangular cross-section, FIG. 3C shows a tube 301 of trapezoidal cross-section with one pair of opposite sides parallel to one another, and FIG. 3D shows a tube 302 of square cross-section. In order to avoid a short coupling path for sound energy between opposite sides of the transducer, the two air columns in each tube must be isolated from each other at the point where the transducer is positioned within the tube. In the case of a tube of circular cross-section, isolation is easily accomplished by employing a transducer with a circular diaphragm, whereas in the case of a tube with a non-circular cross-section such as the trapezium or square cross-section shown in FIGS. 3C and 3D, isolation requires a

battle

32 or 33 in order to adapt a circular diaphragm transducer to the non-circular tube crosssection. For a rectangular cross-section tube, as illustrated by element 300 in FIG. 3B, it is preferred to employ a transducer 30 with an oval or elliptical diaphragm which more closely matches the rectangular cross-section of the tube than a circular diaphragm, thereby to obtain better efficiency in driving the two air columns. Of course, even with an oval diaphragm it is still necessary to employ a batiie 31 in order to isolate the two air columns at the point where the transducer is positioned within the tube.

In FIG. 4 there is illustrated a complete system for reproducing sound from a single sound source, which may be a source of either live or recorded sound. In the case of a live sound source, a transducer 40, for example, a conventional microphone, detects the live sound and converts it into a corresponding lsingle electrical wave. This single electrical wave is delivered by way of adjustable switch S to a suitable amplifier 42 and thence to one or more speakers 44-1 through t4-n. Each of the speakers 44-1 through 44-n is constructed in accordance with the principles described above, as indicated by the two-to-one relationship between the lengths of the air columns on either side of the corresponding transducers 10-1 through lll-n.

Alternatively, the sound to be reproduced may be recorded, in which case a recorded sound source 4l, which may be any one of a number of well known devices such as disc records, magnetic tapes, magnetic wires and the like, is connected to the system. The recorded sound from source 41 is converted into a correspon-ding single electrical wave by transducer 43, for example, a conventional cartridge or other device designed to convert the recorded sound into corresponding electrical variations. By appropriate .setting of switch S the single electrical wave from transducer 43 is supplied to amplier 42 and thence to one or more speakers 44.

As indicated by the broken line between amplifier 42 and the speakers 44, speakers 44 may be located at points that are remote from the other elements of the system shown in FIG. 4. For example, each speaker 44-1 through 44-n may be located in a different room of a hotel or other place of public accommodation while the other elements of the system may be located at a central con- 4trol station on the premises.

While the invention has been described in connection with specific embodiments thereof and in specific uses, various modifications thereof will occur to those skilled in the art without departing from the spirit and scope of the invention as set forth in the appended claims.

What is claimed is:

1. Apparatus for reproducing sound which comprises a hollow duct of selected length open at each end,

said length being selected to reproduce the audible sound spectrum between predetermined high and low frequencies and being at least on the order of three feet,

a transducer for converting a single electrical Signal into sound energy, and

means for positioning said transducer within said duct to divide the interior of said duct into two separate air columns each having its own selected length, one air column on either side of said transducer, so that each motion of said transducer simultaneously produces a compressional pulse in one of said air columns and a rarefactional pulse in the other of said air columns,

wherein the selected length of one of said air columns is substantially twice the selected length of the other of said air columns in order to obtain substantial coincidence between pressure peaks in the sound energy radiated from one end of said duct and pressure nulls in the sound energy radiated from the other end of said duct, and wherein the combined selected lengths of said two air columns are approximately equal to said selected length of said hollow duct.

2. Sound reproduction apparatus which comprises a hollow tube of selected length open at each end, said length being selected to reproduce the audible sound spectrum between predetermined high and low frequencies and being at least on the order of three feet,

an electro-dynamic, direct radiator transducer for converting a monophonic electrical Wave analogue of an acoustical wave into sound energy, said transducer being provided with a diaphragm which is moved in response to variations in said electrical wave analogue to produce compressional and rarefactional air pulses, and

means for positioning said transducer within said tube to divide the interior of said tube into two separate air columns each having its own selected length, one air column on either side of said diaphragm, so that the selected length of the air column on one side of said diaphragm is substantially twice the selected length of the air column on the other side of said diaphragm in order to obtain substantial coincidence between pressure peaks in the sound energy radiated from one end of said tube and pressure nulls in the sound energy radiated from the other end of said tube, wherein each motion of said diaphragm simultaneously produces a compressional pulse in one of Said air columns and a rarefactional pulse in the other of said air columns and wherein the combined selected lengths of said two air columns are approximately equal to said selected length of said hollow tube.

3. Sound reproducing apparatus which comprises ya hollow tube open at each end and having a selected length and a uniform circular cross-section of selected inner diameter, said length being selected to reproduce the audible sound spectrum between predetermined high and low frequencies and being at least on the order of three feet,

an electro-dynamic, direct radiatior transducer for converting a single electrical wave analogue of an acoustical wave into sound energy, said transducer being provided with a diaphragm which is moved in -response to variations in said electrical wave analogue to produce compressional and rarefactional air pulses, wherein said diaphragm is circular and has a diameter not greater than the inner diameter of said tube, and

4l. Apparatus as defined in claim 3 wherein said selected length of said hollow tube is a maximum of about six feet.

5. Apparatus for reproducing Sound which comprises a rst hollow member open at each end and enclosing a first air column having a first selected length,

a second hollow member open at each end and enclosing a second air column having a second selected length, wherein said second selected length of said second air column is substantially twice the first selected length of said first air column, and wherein the combined selected lengths of said first and second hollow members is selected to reproduce the audible sound spectrum between predetermined high and low freruencies and is at least on the order of three feet, an

transducer means for coupling said first hollow member at one of its open ends to one of the open ends of said second hollow member so that each motion of said transducer simultaneously produces a cornpressional pulse in one of said air cloumns and a rarefactional pulse in the other of said air columns, and so that pressure peaks in the second energy radiated from the other open end of one of said hollow members substantially coincide with pressure nulls in the sound energy radiated from the other open end of the other of said hollow members.

6. Sound reproducing apparatus which comprises a hollow tube open at each end and having a selected length and a cross-section other than circular, said length being selected to reproduce the audible sound spectrum between predetermined high and low fre- 9 quencies and being at least on the order of three feet,

an electro-dynamic transducer for converting a single electrical wave analogue of an acoustical wave into sound energy, said transducer being provided with a diaphragm which is moved in response to variations in said electrical Wave analogue to produce compressional and rarefactional air pulses, wherein said diaphragm is shaped to approximate the cross-section of said tube, and

means for positioning said transducer within said tube to divide the interior of said tube into two separate air columns each having its own selected length, one air column on either side of said diaphragm, so that the selected length of the air column on one side of said diaphragm is substantially twice the selected length of the air column on the other side of said diaphragm in order to obtain substantial coincidence between pressure peaks in the sound energy radiated from one end of said tube and pressure nulls in the sound energy radiated from the other end of said tube, wherein each motion of said diaphragm is simultaneously produces a compressional pulse in one of said air columns and a rarefactional pulse in the other of said air columns, and wherein the combined selected lengths of said two air columns are approximately equal to said selected length of said hollow tube.

v7. Apparatus as defined in claim 6 wherein said tube cross-section is rectangular, the shape of said diaphragm is oval, and said positioning means includes a baffle to isolate said two air columns from one another.

8. Apparatus as defined in claim 6 wherein said tube cross-section is square, the shape of said diaphragm is circular, and said positioning means includes a baffle to isolate said two air columns from one another.

9. Apparatus as defined in claim 6 wherein said tube cross-section is trapezoidal with at least one pair of opposite sides parallel to one another, the shape of said diaphragm is circular, and said positioning means includes a bafe to isolate said two air columns from one another.

.10. A stereophonic sound reproduction system which comprises a source of sound,

first transducer means responsive to said source of sound for deriving from said source of sound a single electrical wave representative of said sound,

amplifier means connected to said first transducer means, and

n speaker means in circuit relation with said amplifier Imeans for converting said electrical wave into lsound energy, where n is a selected positive integer equal to or greater than one, each of said speaker means including a corresponding one of n second transducer means positioned within a hollow tube of selected length, said selected length being at least on the order of three feet, and said tube being open at both ends to divide the interior of said tube into two separate air columns each having its own selected length, one air column on either side of said second transducer, so that each motion of said second transducer simultaneously produces a compressional pulse in one of said air columns and a rarefactional pulse in the other of said air columns, and so that pressure peaks in the sound energy radiated from one end of said tube substantially coincide with pressure nulls in the sound energy radiated from the other end of said tube, wherein the length of said hollow tube is selected to reproduce the audible sound spectrum between predetermined high and low frequencies, wherein the selected length of one of said air columns is substantially twice the selected length of the other of said air columns, and wherein the combined selected lengths of said two air columns are approximately equal to said selected length of said hollow tube.

References Cited UNITED STATES PATENTS 2,058,132 10/1936 Cirelli 181-27 2,896,737 7/1959 Gellman 181-31 2,905,259 9/1959 Ashe 181-27 3,080,785 3/1963 Evans 181-31 X FOREIGN PATENTS 143,597 9/1951 Australia.

STEPHEN I. TOMSKY, Primary Examiner.