US2127672A - Sound attenuating device - Google Patents

Description

Aug. 23, 1938. R. B. BOURNE i SOUND ATTENUATING DEVICE Filed Oct. 25, 1935 MJMQBMQ. RKKY n @a lr- I/ INVENTOR R0/.M0 AZ300/PMS J- BY l ToR' NEYs Patented Aug. 23, 1938 UNITED STATES PATENT OFFICE SOUND ATTENUATING DEVICE Roland B. Bourne, Hartford, Conn., assignor to The Maxim Silencer Company,

Hartford,

4 Claims.

The present invention relates to improvements in sound attenuating devices employing reactive acoustic sidebranches acoustically coupled to a main sound conducting channel.

The types of acoustic sidebranches employed in the several embodiments of the invention are classed as linear sidebranches, wherein change of acoustic phase takes place as a function of distance and wherein the length of the sidebranch is ordinarily a principal factor in determining its resonance frequencies. The invention makes use of various types of linear sidebranches which may be acoustically coupled to a rnain sound conducting channel or other enclosure. This acoustical coupling may be accomplished in different manners.

A characteristic of linear resonators in general is the fact that they have a plurality of modes of vibration, the natural frequencies forming a series. In the case of a linear resonator of ccnstant cross sectional area throughout its length, and having unrestricted openings at its end or ends, and in the case of a complete cone entirely open at its base and closed at its vertex, these natural frequencies form an integral series. In the case of certain other types of linear resonators having unrestricted openings, the natural frequencies do not form an integral series.

A linear resonator acoustically coupled to a main sound conducting channel offers attenuation to sound waves, in the channel, of frequencies directly dependent upon the natural frequencies of the resonator. It is obvious that a resonator having an integral series of response frequencies may be used to attenuate a sound Wave having an integral series of component frequencies. Two or more resonators may be associated with a main sound conducting channel in such a way that substantially continuous attenuation over a wide range of frequencies may be obtained. 'Ihe sound attenuating device thereby resulting may take the form of a sound conducting channel having a system of discrete resonators acoustically coupled to it or it may take the form of an acoustic Wave lter having recurrent pass and attenuation bands or there may result various combinations whereby a desired acoustic result is obtained.

Under certain conditions, the length of acoustlc sidebranches of the linear type may be a disadvantage from the viewpoint of available space for an installation. This is especially true in the case where the objectionable sound Waves are of extremely low frequency. Any means for ap- L preciably altering its lowest response frequency is desirable. This is one of the main purposes of the invention.

Another important purpose of the invention is to provide means for converting a non-integral series of response frequencies of a linear type resonator into a substantially integral series. A further purpose of the invention is to provide means whereby an acoustic sidebranch may be designed having a non-integral series of response frequencies so that attenuation may be offered to corresponding non-integrally related sound frequencies passed by an acoustic wave lter or other selective sound attenuating system. Further purposes and objects of the invention will be disclosed as the specification proceeds.

The purposes of the invention are accomplished by the use of a restricted opening or its equivalent at the mouth of the linear acoustic sidebranch, at the point where the sidebranch is acoustically coupled to the main sound conducting channel or other enclosure wherein objectionable sound waves may occur.

The use of a restricted opening at the mouth of a linear resonator, which is at least a quarter wave length long, is not comparable to the use of a restricted opening giving into a volumetric or Helmholtz type resonator, which is small compared to a wave length. In the latter case, the acoustic conductivity of the opening is a principal factor in determining the resonance frequency of the resonator, the frequency varying directly as the square root of the conductivity of the opening. The volumetric resonator, being of small dimensions relative to a wave length of sound for its resonance frequency, does not depend upon any particular dimension for computing its acoustical capacitance or volume. For instance, if of cylindrical shape, the volume and therefore the capacitance of the resonator might be held constant whilst varying both the length and diameter appropriately, the resonance frequency remaining constant for a given medium and a given conductivity. In the case of a linear resonator of cylindrical shape, however, the volume may be, for instance, doubled, Without appreciably changing the fundamental frequency of response as long as the length is unaltered. The acoustic system comprising a linear resonator with a restricted coupling opening may be said to be a concentrated or lumped inertance in series with a distributed inertance and capacitance.

In the design of linear resonators used as acoustic sidebranches, the conductivity of the mouth has been generally taken into account by assuming that the sidebranch extends a certain distance into the channel or enclosure to which it is coupled. This procedure results in an effective increase, generally quite small, in the length of the sidebranch, the ratio between the various modes of vibration remaining unchanged by the process.

In my invention, however, the use of a substantially restricted openingr between the linear resonator proper and the main sound conducting channel not only increases the acoustic length of the sidebranch by a very appreciable amount, but also materially alters the relation between the fundamental or lowest response frequency and the several overtones, Whether these overtones be harmonic or not. There is a practical limit to the extent that the conductivity may be reduced. In the various embodiments of my invention, the conductivity is formed by the most eiiicient orifice possible, so that no appreciable additional amount of acoustic resistance due to surface friction, eddies, narrow openings and the like is introduced. Therefore a preferred type of opening to form the necessary conductivity is a single round hole in a flat plate. In such embodiments as require slot-like conductivities, such openings are disposed in centrally located main conducting channels rather than` in annular channels having large peripheries and therefore higher acoustic resistance. The acoustic efficiency of a sidebranch is impaired by the introduction of acoustic resistance thereinto.

The invention finds useful application in sound attentuating devices or silencers for use in conjunction with internal combustion engine eX- hausts, air intakes, discharges and the like. In commercial embodiments, the over-all length of the silencer is directly influenced, in many cases, by the lengths of the acoustic sidebranches, since they are usually disposed parallel to the axis of the main channel through the device; and therefore, substantial diminution of the lengths of the sidebranches is reiiected directly in the increased commercial adaptation of the silencer due to its shorter length.

Inasmuch as some of the embodiments herein shown employ acoustic sidebranches of non-uniform cross sectional area as a function of distance along their lengths, I refer to my Patents 2,017,745; 2,017,746; 2,017,747; and 2,017,748, granted October 15, 1935, wherein several such acoustic sidebranches are shown.

Referring to the drawing,

Fig. 1 shows a cylindrical resonator embodying the principle of the invention;

Fig. 2 shows a conical resonator embodying the principle of the invention;

Fig. 3 shows a truncated conical resonator embodying the principle of the invention;

Fig. 4 shows a sound attenuating device illustrating an application of the invention;

Fig. 5 shows a graph germane to Figs. 4 and 6;

Fig. 6 shows a composite sound attenuating device embodying the principle of the invention; and

Fig. 7 shows a simple sound attenuating device employing the resonators of Fig. l and Fig. 2.

The resonator of Fig. l comprises the cylindrical casing I, of uniform cross sectional area throughout its length, closed at one end by the transverse header 2 and fitted at the other end by the transverse plate 3 having a centrally disposed opening 4 forming an acoustic conductivity Co. This discussion is limited to embodiments wherein the length L of the resonator is appreciably greater than its diameter D and wherein the length L is an appreciable fraction of a wavelength for the lowest sound frequency to which the resonator resonates. If L is equal to D, for instance, the device functions as a volumetric or Helmholtz resonator. While there is no sharp line of demarcation between the Helmholtz and so-called linear resonator, the resonators involved in my invention are sufficiently long as to be adequately treated on the basis of distributed acoustic elements of inertance and capacitance rather than on the lumped theory, which is applicable to the Helmholtz type. The resonator of Fig. 1, with the end plate 3 removed has response frequencies determinable from wL Il Tf-.5, 1.5, 2.5, etawhere n 1s odd (1) ci: 21rx frequency L=acoustc length of the resonator including an end correction C=vclocity of sound in the medium The addition of theI end plate 3 with a suitable aperture 4 therein materially alters this series of response frequencies. A representative series thus obtained is given by relation has been converted into a non-integral one. In terms of the length of the device to respond to a given frequency, it is seen that a saving of 20% is effected. Such a resonator, when used as an acoustic sidebranch acoustically coupled to a sound conducting channel will offer attenuation to sound frequencies determinable in accordance with an equation similar to Equation 2.

Fig. 2 shows a complete conical resonator of slant length L and diameter D. The cone 5 is fitted at its large end with the transverse end plate 6 having a circular opening 1 therein to form a conductivity Co. Without the end plate 6, the conical resonator has natural frequencies which may be shown to be given by ;:=1, 2, 3, 4, 5, etc. (3)

Such resonators are discussed in my patents above referred to. The addition of the end plate 6 with its acoustic conductivity Co alters the above series of response frequencies. A representative series is shown by It will be seen that the effect of the conductivity Cu in the case of the cone is very similar to its effect in the case of the cylindrical resonator of Fig. 1.

Certain linear type resonators having unrestricted openings thereinto have response frequencies which do not bear an integral relation to each other. An example of a linear type resonator having a non-integral series of response frequencies is a truncated cone, open at its base and closed at its small end by an imperforate header. Fig. 3 shows a truncated conical resonator of length L and diameter D comprising the tapered shell 8, closed at its small end by the n R-. 1.56, 2.54, 3.52, which approaches as a limit where n is an odd integer.

Equation (5) shows that this type of resonator yields av series of response frequencies resembling those obtained for a closed cylinder of uniform cross sectional area and having no restricted opening thereinto, as given by Equation (l), eX- cept that the fundamental frequency is higher and succeeding overtones are also higher but by a decreasing amount approaching a limit for the nth overtone, Where the resonance frequencies of the tWodevices coincide. In other words, for high overtones, the truncated cone behaves like a closed cylinder. By the use of the end plate II) having an aperture II therein, I have found that it is possible and readily practicable to choose a value for the conductivity C0, as represented by the aperture II, whereby the device yields substantially the same integral series of response frequencies as 4does a closed, cylindrical resonator of uniform cross sectional area having no restricted opening at its mouth, said frequencies being given by Equation (l) above. Different ratios between the diameters D and d require different conductivities in order to bring the response frequencies into an integral series. For instance, for the case where D=3d', the value of Co should be less than for the case where D=2d. The conductivity Co is preferably formed by a single aperture, as shown, rather than a number of smaller apertures, in order to keep the acoustic resistance ofthe opening low. The truncated conical resonator of Fig. 3 could be used .as an acoustic sidebranch in association with a main sound conducting channel to attenuate therein a series of sound frequencies as given by Equation (1). There would result a saving in total volume over a cylindrical sidebranch of the same frequency response.

A simulation of the truncated cone of Fig. 3 is shown incorporated in the device of Fig. 4. This device is a sound attenuator having entirely novel features. It comprises the generally cylindrical casing I2, having centrally disposed openings I3, IlI at either end thereof, and a unit coaxially mounted therewithin comprising the cylindrical shell I5, the cone I 6, having its open base afxed to one end of the cylinder i5 and extending inwardly to a point in adjacency to the other end of said cylinder I5, whereby are formed the conical sidebranch i? and the conico-annular sidebranch I8. The open end of the cylinder l5 is fitted with the transverse header I9 having a centrally disposed aperture 20 therein. Both the conical sidebranch I'I and the conico-annular sidebranch i3 have an acoustic length L, as shown. These two sidebranches are acoustically coupled to opposite ends of an annular main sound conducting channel M which is formed between the inside of the casing I2 and the outside of the cylinde-r I5. The sidebranch unit is, of course, made shorter than the casing I2 so that the main channel (and path for exhaust or other gases) extends from the opening I3 to the opening I4 as shown by successive radial and annular paths. I have found that it is readily possible to choose a value for the conductivity of the aperture 20 whereby the conico-annular sidebranch i8 offers peak attenuation to a series of sound frequencies which do not depart sensibly from those given by Equation (l) above. Since, as has been pointed out, the conical sidebranch I'I offers peak attenuation to sound frequencies given by Equation (3), it is readily seen that the device of Fig. 4 offers peak attenuation to a series of sound frequencies given by mL lT-.5, 1.0, 1.5, 2.0, 2.5, etc.

These frequencies are the same as would be given by a single cone of acoustic length 2L. In the latter case, however, the attenuation would be theoretically zero for a series of intermediate frequencies, While in the case of the device of Fig. 4 the attenuation does not fall to Zero at any intermediate frequency. Fig. 5 shows a plot of measured attenuation vs. frequency characteristics for a device built in accordance with Fig. li, and having an acoustic length of L. It will be seen that the actual attenuation peaks occur at frequencies substantially in accordance with Equation (6) Another set of abscissae are given for the case Where the acoustic length is %L, as discussed later. A similar type of curve is obtainable for a cone and cylinder used as successive sidebranches along a main conducting channel, but such a device would be twice as long as that shown in Fig. 4. The practical utiiity of this embodiment of the invention is in providing a simple, cheap and effective attenuator for a simple sound having overtones in harmonic relation.

A further useful purpose of the device of Fig. i is shown in Fig. 6 wherein one unit of a composite sound attenuating device is designed to offer attenuation to sound frequencies suffering little or no attenuation` in their passage through another part thereof. It comprises the generally cylindrical casing 2| divided by the transverse header 22 into two sections, A and B, respectively. Section A comprises the main sound conducting channel 23 formed by the tubular member 24 extending longitudinally therethrough and the closed cylindrical paired acoustic sidebranches 25, 26, acoustically coupled to the main channel 23 through the slots 2'1, 28 respectively. The relative dimensions of the sidebranches with respect to the distance along the main channel between them are shown in the figure. This device is completely described in United States Patent No. 1,910,672. This section A is an acoustic Wave filter having a series of attenuation and pass bands. The pass bands are at values of These Values are seen to be in the proportion 1:2:324: etc. Consequently the device of Fig. 4 is readily adaptable for providing peak attenuation to those sound frequencies represented by Equation (7). Accordingly, section B of Fig. 6 contains the sound attenuating unit 29 of acoustic length %L, as shown. Data for Fig. 5 was taken With the device inserted in an acoustically long line and what is plotted is the insertion, loss in decibels.

A sound attenuating device which provides peak attenuation for a non-linear series of sound frequencies in two overlapping characteristics is shown in Fig. '7. It comprises the generally cylindrical casing 30 divided into two compartments by the perforate transverse header 3l. In one of said compartments is mounted the closed cylindrical sidebranch 32 constructed after the manner of Fig. 1. In the other of said compartments is mounted the conical acoustic sidebranch 33 constructed after the manner of Fig. 2. In addition, a simple system of baffles 34 for additional high frequency attenuation is mounted exteriorly of the cone forming the conical sidebranch 33. Each of the two acoustic sidebranches contributes to the total attenuation in a manner similar to that indicated in Equations 2 and 4.

While I have shown but a few of the applications of the principle of my invention to practical embodiments thereof, many more and useful applications may be made within the scope of this disclosure.

I claim:

1. A sidebranch structure adapted for incorporation in a sound conducting channel which comprises a cylindrical shell, a conical shell telescoped wi. hin the cylindrical shell and joined thereto at one end, whereby a conical sidebranch is formed within the conical shell and a conico-annular sidebranch is formed between the conical and the cylindrical shells, both of said sidebranches being open at their larger ends, and an apertured plate partially closing the opening into the conico-annular sidebranch.

2. A sidebranch structure adapted for incorporation in a sound conducting channel which comprises a Cylindrical shell, a conical shell telescoped within the cylindrical shell and joined thereto at one end, whereby a conical sidebranch is formed Within the conical shell and a conicoannular sidebranch is formed between the conical and the cylindrical shells, both of said sidebranches being open at their larger ends, and an apertured plate partially closing the opening into the conico-annular sidebranch, the opening in the apertured plate being suiliciently restricted to cause the conico-annular sidebranch to respond to a series of frequencies substantially intermediate those to which the conical sidebranch responds.

3. An acoustic silencing device comprising a tubular member defining the outer boundary of a main sound conducting channel, a cylindrical member mounted Within the tubular member and spaced therefrom so as to define throughout its length the inner boundary of the main sound conducting channel, a conical member telescoped within the cylindrical member and joined thereto at its larger end, whereby a conical sidebranch is formed within the conical shell and a conicoannular sidebranch is formed between the conical and the cylindrical shells, both of said sidebranches being open at their larger ends, and an apertured plate partially closing the opening into the conico-annular sidebranch.

4. An acoustic silencing device comprising a tubular member defining the outer boundary of a main sound conducting channel, a cylindrical member mounted within the tubular member and spaced therefrom so as to deiine throughout its length the inner boundary of the main sound conducting channel, a conical member telescoped within the cylindrical member and joined thereto at its larger end, whereby a conical sidebranch is formed within the conical shell and a conicoannular sidebranch is formed between the conical and the cylindrical shells, both of said sidebranches being open at their larger ends, and an apertured plate partially closing the openingr into the conico-annular sidebranch, the opening in the apertured plate being suciently restricted to cause the conico-annular sidebranch to respond to a series of frequencies substantially intermediate those to which the conical sidebranch responds.

ROLAND B. BOURNE.

Til

Images