My invention relates to. electroacoustic transducer arrangements, and more particularly, to electroacoustic transducer filtering assemblies for use in digital communication systems. Background of the Invention

In digital communication systems, intelligence is generally conveyed in the form of sequential pulse codes. The transmission of an audio signal over a digital facility requires that the audio signal be sampled at a rate greater than twice the highest frequency to be transmitted. All frequency components of the audio signal and noise above one-half the sampling frequency must be suppressed prior to sampling. Otherwise, the modulation introduced by the sampling process results in a folded audio signal spectrum which appears as interference in the frequency range below one-half the sampling frequency. This effect, generally known as aliasing, is prevented by accurately controlling the passband of the audio signal prior to sampling. While aliasing only occurs at transmitting terminals of a digital communication system where an audio signal is sampled, similar passband control arrangements are needed at receiving terminals of the digital facility. Otherwise, the modulation introduced by sampling results in- an interferring audio signal above one-half the sampling frequency being applied to the receiving transducer. Electrical filters designed to prevent interference in- -.-..-- digital sampling systems are complex and expensive since they must provide accurate filtering over the entire audio

2. receiver cover. The cover is shaped so that sound waves from the transducer are directed through a pair of cavities dimensioned to attenuate higher frequency components. A Helmholtz resonator is incorporated in the receiver cover 5 to provide absorption of sound at its resonant frequency. The Hel holz cavity resonant frequency is set just below one-half the sampling frequency of the digital facility to which the transducer is connected. In this way, the filtering required for digital communication of voice

10 signals is achieved.

The Baechtold scheme as well as other prior art acoustic filter networks (e.g., U. S. Patent 3,452,164 ^' issued June 24, 1969 to Kiyoshi Kobara) are adapted to modify the frequency characteristics of sound waves

15 travelling along the axis of the acoustic network associated with a telephone receiver. Such arrangements are practical as long as structural dimensions of the acoustic network are small in comparison with the wavelengths of the sound passing therethrough. When the

20 sound wavelengths are comparable to the structural dimensions, as is generally the case in telephone type transducers and loudspeakers, the higher frequency sound waves cause resonance in the acoustic network. Consequently the unwanted sound wave frequencies are not

25 adequately attenuated by the acoustic filter.

In analog communication systems, higher frequency sound waves outside the desired passband cause little difficulty since they are readily removed in the system. In digital communication systems utilizing sampling,

30 however, the sound wave components of frequency greater than one-half the sampling frequency whose wavelengths are comparable to the acoustic network dimensions partially pass through the network. As aforementioned, these unwanted components are folded into the desired signal

35 frequency band and cause serious interference at transmitter terminals or are passed through the acoustic network associated with the receiving transducer at receiver terminals.

Description of the Drawing

FIG. 1 depicts an exploded view of a cylindrical transducer assembly illustrative of the invention;

FIG. 2 shows a sectional view of the transducer assembly of FIG. 1;

•FIG. 3 shows an equivalent electrical circuit for the transducer assembly of FIGS. 1 and 2;

FIG. 4 depicts an exploded view of a modified cylindrical transducer assembly illustrative of the invention;

FIG. 5 shows waveforms illustrating the frequency responses of the transducer assemblies of FIGS. 1, 2, and FIG. 4;

FIG. 6 shows an equivalent electrical circuit of the modified cylindrical transducer assembly of FIG. 4;

FIG. 7 depicts an exploded view of a rectangular transducer assembly illustrative of the invention;

FIG. 8 shows an equivalent electrical circuit of the rectangular transducer assembly of FIG. 7; and FIG. 9 shows waveforms illustrating the frequency response of the rectangular transducer assembly of FIG. 7. Brief Summary of the Invention

The invention is directed to an electroacoustic assembly that includes a housing having an aperture plate end, an open end and a longitudinal passageway of predetermined length and cross section. The open end of the housing is securable to a transducer and a plurality of apertured plates are secured in the housing at spaced intervals along the passageway. Each spaced interval includes structural elements that partition the spaced interval into a plurality of longitudinal sections to inhibit resonance or the housing. The spaced intervals, the apertures in the plates, and the structural elements are dimensioned relative to the passageway cross section to suppress passage of sound waves outside a predetermined frequency band through the passageway.

According to one aspect of the invention, t

orientation of the partitioned sections of each spaced interval is offset from the orientation of the adjacent spaced interval partitioned sections.

According to another aspect of the invention, the partitioned sections of at least one spaced interval further includes a tubular chamber which may have a cross-section other than circular secured between adjacent plates at apertures therein. The chamber sidewall has apertures opening in the partitioning sections.

According to yet another aspect of the invention, the electroacoustic transducer assembly includes a transducer containing end plate and a plurality of tubular members. Each tubular member comprises an apertured plate end, an open end and a tubular cavity of predetermined cross section. The tubular members are tandemly arranged with the open end of a tubular member being secured to the • plate end of the adjacent tubular member to form a housing having a divided longitudinal passageway therethrough. Each tubular cavity is partitioned into longitudinal sections by structural elements to inhibit resonance of the member. The apertures, the lengths of tubular cavities, and the structural elements are dimensioned relative to the tubular cavity cross sections to suppress passage of sound waves outside a predetermined frequency band through said passageway.

According to yet another aspect of the invention, the transducer assembly forms a part of a digital communication system having a predetermined sampling frequency. The apertures in the plate ends of the tubular members, the lengths of the tubular cavities, and the structural members are dimensioned relative to the tubular member cross sections to allow substantially uniform passage of sound waves below one-half the sampling frequency of the digital facility and to suppress passage of sound waves at and above one-half said sampling frequency through said passageway. The resonant chambers and apertures therein are dimensioned to

at one-half the sampling frequency. Detailed Description

FIG. 1 depicts an exploded view of an electroacoustic transducer assembly illustrative of the invention in which apertured plate 10 is secured to end 2 of cylindrical housing 1. Structural cross members 15 and 17 are attached to the right side of apertured plate 10. Plate 10 is divided into four sectors by cross member structural elements 15 and 17, and each sector has at least one aperture, e.g., 12-1, 12-2, 12-3, and 12-4. In like manner, apertured plate 20 is divided into four sectors by structural elements 25 and 27, and apertured plate 30 is similarly divided by cross members 35 and 37. Plate 20 is inserted into housing 1 so that its left side contacts structural cross members 15 and 17, and plate 30 is inserted so that its left side contacts structural members 25 and 27. Transducer 5, which may be a microphone, a telephone receiver, or a loudspeaker, is secured to end 3 of housing 1. The electroacoustic transducer assembly of FIG. 1 is thereby divided into a series of three cavities, the dimensions of which are determined by the structural elements and the cross section of housing 1. Each cavity is partitioned into four longitudinal sectors by the structural elements therein. For example, the cavity between plates 10 and 20 is divided into sectors 14-1, 14-2, 14-3 and 14-4 by structural elements 15 and 17, and the cavity length is determined by the width of rectangular cross members 15 and 17. It is well known in the art that cross mode resonances may occur in acoustic cavities when the wave lengths of sound waves applied thereto are comparable to the transverse dimensions of the cavity. The diameter of a standard telephone handset microphone or receiving device is such that the cavity resonances therein are well within the audio range. The longitudinal partitioning of the cavities by structural elements in accordance with the invention is operative to inhibit resonances in the audio

OMPI

range so that the aforementioned sampling modulation effects occurring in digital facilities are avoided. -In the arrangement of FIG. 1, plate 10 is oriented in housing 1 so that structural member 15 is horizontal and structural member 17 is vertical. Plate 20 is rotated clockwise with respect to plate 10 whereby the orientation of structural members 25 and 27 is offset from the orientation of structural members 15 and 17 and apertures 22-1, 22-2, 22-3 and 22-4 are free of interference from structural members 15 and 17. In similar manner, plate 30 is rotated clockwise with respect to plate 20 whereby the orientation of structural members 35 and 37 is offset from the orientation of members 25 and 27 and apertures 32-1, 32-2, 32-3 and 32-4 are out of-line with structural members 25 and 27.

The different orientations of the structural members attached to plates 10, 20 and 30 are effective to further inhibit the resonance of housing 1 responsive to sound waves of wavelengths comparable to the diameter of the housing. The offset relationship of the apertures in plate 10, 20 and 30 insures that each cavity in housing 1 is operative independently of the other cavities to control the sound waves passing through the housing. Although the structural elements in FIG. 1 are pairs of perpendicular members, it is to be understood that other structural element configurations provide similar results. For example, a greater number of structural elements may be used or curved structural elements may be employed. Further, the aperture arrangement of FIG. 1 may be modified to include a plurality of apertures in each cavity sector so long as the adjacent structural elements do not intersect the apertures.

FIG. 2 shows a cross section of the assembled transducer arrangement depicted in FIG. 1 taken through line 2-2 where transducer 5 is a cylindrical microphone adapted to receive sound waves from sound source 50. The sound waves enter sectors 14-1, 14-2, 14-3 and 14-4 of the cavity between plates 10 and 20 through apertures 12-1,

12-2 , 12-3 and 12-4 . Each aperture exhibi ts an acoustic i nertance corresponding to ρ_ L = d

5 where P is the air density and d_ is the aperture diameter.

The aperture also exhibits a loss factor R associated primarily with the viscous loss at the wall of the aperture. Additional viscous loss may be provided by

10 covering the aperture with a silk screen or cloth. The value of the loss factor R in the arrangement of FIGS. 1 and 2 is determined experimentally as is well known in the art.

Each sector of the cavity between plates 10 and

15 20 is effective as an acoustic compliance

Al

C a pC

where A is the cross section area of the sector, _ is the

90 axial length of the cavity and £ is the velocity of sound. A conductance G is representative of the loss in the sector which is primarily associated with the heat conductance of the sector walls. The conductance G is determined experimentally for the configuration of FIGS. 1 and 2 as is

25 well known in the art.

Sound waves from sources 50 are modified by passage through apertures 12-1, 12-2, 12-3 and 12-4, and cavity sectors 14-1, 14-2, 14-3 and 14-4. The modified sound waves then enter cavity sectors 24-1, 24-2, 24-3 and

30 24-4 between plates 20 and 30 via apertures 22-1, 22-2, 22-3 and 22-4, and are altered therein in accordance with the aperture and cavity dimensions. Alter passage through apertures 32-1, 32-2, 32-3 and 32-4 in plate 30, and cavity sectors 34-1, 34-2, 34-3 and 34-4, the further modified

35 sound waves are applied to microphone 5. The apertures in plates 10, 20 and 30, the cavity sector lengths and the structural elements of FIGS. 1 and 2 are dimensioned relative to the cross section of housing 1 so that passage

U L-.1

OMPI

of sound waves of frequencies outside a predetermined band is suppressed.

FIG. 3 shows an electrical circuit equivalent of the acoustic network of FIGS. 1 and 2. In FIG. 3 resistances 12', 22* and 32' represent the equivalent acoustic losses of the apertures in plates 10, 20 and 30, respectively. Inductances 12'*, 22' ' and 32* * represent the equivalent acoustic inertances of the apertures in plates 10, 20 and 30, respectively. Capacitance 14' represents the equivalent acoustic complicance of cavity sectors 14-1, 14-2, 14-3 and 14-4, while conductance 14' ' corresponds to the combined acoustic losses of the sectors. In similar manner, capacitance 24' represents the combined acoustic compliance of cavity sectors 24-1, 24-2, 24-3 and 24-4, and capacitance 34' represents the equivalent acoustic compliance of cavity sectors 34-1, 34-2, 34-3 and 34-4. Conductances 24* * and 34' * represent the equivalent losses of sectors 24-1, 24-2, 24-3 and 24-4, and sectors 34-1, 34-2, 34-3 and 34-4, respectively. As is well known in the art, the circuit of

FIG. 3 is of the uniform ladder type and its transfer function is

-L P 1.. x ³ +5x ² +6x+l

where X = - ^z l ^z 2 ^z l - R+j L and ² 2 = 1 G+j C

R is the equivalent.loss of the apertures of a plate;

L is the equivalent inertance of the apertures of the plate;

C is the equivalent compliance of the sectors of a cavity;

G is the equivalent loss of the sectors of a cavity;

P^ is the sound pressure at the transducer; and P^ is the sound pressure at plate 10.

< OM /,- WIP

• aveform 502 of FIG. 5 illustrates the frequency response of a transducer assembly constructed in accordance with FIGS. 1 and 2 for use in a digital facility where transducer 5 is an electret microphone having a diameter of 46 millimeters and the sampling frequency is 8 kilohertz. The equivalent ladder circuit of FIG. 3 for the sectored cavity arrangement of FIG. 1 includes inductances of 1.61 x 10 ^J cgs units, capacitances of 4.0 x 10 cgs units, aperture loss resistances 7.5 cgs units and. negligible cavity loss conductances. These parameters represent the parallel connection of the four partitioned cavity sectors. The apertures in FIG. 1, to obtain the desired inductances and resistances are 4.5 millimeters in diameter and the cavity lengths to obtain the desired capacitances are 4.3 millimeters. As shown in waveform 502 of FIG. 5, the frequency response is generally uniform up to 3.5 kilohertz, falls off to a low value at 4 kilohertz and remains at the low value in the remainder of the audio spectrum. Waveform 501 of FIG. 5 illustrates the frequency response obtained for the transducer assembly of FIGS. 1 and 2 where the structural members e.g. 15, 17, 25, 27, 35, 37 of FIG. 1, in the cavities are omitted. Waveform 501 exhibits a distinct unwanted response in the range of 8 to 10 kilohertz which response results from the aforementioned cross mode resonance of the plates and the housing. In accordance with the invention, the higher frequency resonances caused by sound waves of wavelength comparable to the 46 millimeter microphone diameter are inhibited by the inclusion of differently oriented cavity cross member structural elements.

FIG. 4 shows an exploded view of another transducer assembly illustrative of the invention. In FIG. 4 cylindrically shaped tubular members 100, 102 and 104 are tandemly arranged to form a housing with a divided passageway therethrough. Tubular member 100 includes apertured plate end 110, cylindrical wall 111 and open end 113. Perpendicularly crossed rectangular members 115 and 117 attached to plate end 110 and wall 111 partition

the cavity of tubular member 100 into sectors- 114-1, 114- 2, 114-3 and 114-4. Each sector of plate end 110 includes an aperture centered in the sector, e.g., aperture 112-1 is centered within sector 114-1. Similarly, tubular member 102 5 has an apertured plate end 120, a cylindrical wall 121 and an open end 123. Crossed rectangular members 125 and 127 ^■ partition the cavity of tubular member 102 into four sectors, 124-1, 124-2, 124-3 and 124-4. Sound waves enter each of the sectors via an aperture, e.g., sector 124-1 is 0 entered via aperture 122-1.

Tubular member 104 comprises apertured plate end 130, cylindrical wall 131 and open end 133. Crossed rectangular elements 135 and 137 partition the cavity of tubular member 104 into sectors 134-1, 134-2, 134-3 and 5 134-4. An aperture is included in each sector and a cylindrical chamber haying one end surrounding the sector aperture extends the length of the cavity from plate end 130 to transducer 105. Chamber 138-1 in sector 134-1, for example, includes end 141 attached to plate 130 around

20 the aperture in sector 134-1. Cylindrical wall 143 extends from plate 130 to transducer 105 and has four equally spaced apertures therein, e.g., aperture 142, which apertures communicate between the chamber cavity within wall 143 and the cavity sector 134-1. Sectors 134-2, 134-

253. and 134-4 also include apertures chambers 138-2, 138- 3 and 138-4, respectively. The construction of each of these chambers is substantially similar to the construction of chamber 138-1. Each chamber and the apertures therein are dimensioned to resonate at a predetermined frequency,

30 i.e., one-half the sampling frequency of the associated digital facility.

The transducer arrangement of FIG. 4 is assembled by securing open end 113 of tubular member 100 to plate end 120, securing open end 123 of tubular member 102 to

35 plate end 130 of tubular member 104, and securing open end 133 of tubular member 104 to the periphery of transducer 105. The tubular members and the transducers are positioned so that there is a common

As described with respect to FIGS. 1 and 2, the orientation of the rectangular structural elements of each plate is offset to inhibit cross mode resonances of the tubular members. In FIG. 4, rectangular structural element 115 5 is horizontal and rectangular structural element 117 is vertical. The orientation of tubular member 102 is rotated clockwise with respect to tubular member 100 whereby the structural elements and the apertures of plates 100 and 120 are offset. Similarly, the

10 orientation of tubular member 104 is rotated clockwise with respect to tubular member 102 whereby the structural elements and apertures of plate 130 are offset from the structural elements and apertures of plate 120.

Sound waves enter cavity sectors 114-1 through

15114-4 via apertures 112-1 through 112-4, and are modified in accordance with the dimensions of the cavity sectors of tubular member 100. The modified sound waves then enter cavity sectors 124-1 through 124-4 via apertures 122-1 through 122-4 and are altered in tubular member 102. The

20 sound waves from tubular member 102 enter cylindrical resonant chambers 138-1 through 138-4 via the apertures in plate 130. A portion of the sound waves in the chambers is applied directly to transducer 105. Another portion of the sound waves enters sectors 134-1 through

25134-4 via the apertures in the tubular chambers and are applied to transducer 105 via cavity sectors 134-1 through 134-4.

FIG. 6 shows the equivalent electrical circuit for the acoustic network of FIG. 4. In FIG. 6

30 resistance 112' and inductance 112' * represent the combined viscous loss and inertance of apertures 112-1 through 112- 4. Capacitance 114' and conductance 114' ' represent the combined compliance and loss of cavity sectors 114-1 through 114-4. Resistance 122' and inducatance 122' ' is

35 equivalent to the combined viscous loss and inertance of apertures 122-1 through 122-4, while capacitance 124-' and conductance 124' ' are equivalent to the combined compliance and loss of cavity sectors 124-1 through 124-4. Similarly,

resistance 132* and inductance 132' * represent the combined viscous loss and inertance of the apertures in plate end 130. Capacitance 134' and conductance 134' ' represent the combined compliances and losses of cavity sectors 134-1 through 134-4, excluding chambers 138-1 through 138-4.

The combined inertances of the apertures in the chamber walls of tubular chambers 138-1 through 138-4 are represented by inductance 142* and the combined compliances of chambers 138-1 through 138-4 are represented by capacitance 141'. Capacitance 141* and inductance 142' form a series resonance circuit having a resonant frequency of one-half the sampling frequency of the associated digital communication system. This resonance circuit is in parallel with capacitance 134* in FIG. 6. The frequency response of the transducer assembly of FIG. 4 for a cylindrical microphone having a diameter of 46 millimeters and a sampling frequency of 8 kilohertz is shown in waveform 503 of FIG. 5. As mentioned with respect to waveform 502, the inclusion of differently oriented structural elements in the cavities of tubular elements 100, 102 and 104 are operative to inhibit resonance of the assembled housing of FIG. 4 so that sound waves of frequencies above 4 kilohertz are suppressed. The inclusion of a plurality of cylindrical chambers 138-1 through 138-4 in the cavity sectors between plate end 130 and transducer 105 modifies the frequency response of the transducer assembly in the region of one-half the sampling frequency whereby a much sharper cutoff characteristic is obtained for the acoustic filter network. While the transducer assembly of FIG. 4 utilizes resonant chambers in one tubular cavity, it is to be understood that resonant chambers may be included in other tubular members as well.

FIG. 7 shows an exploded view of a transducer assembly illustrative of the invention which is adapted for use with a rectangular transducer 205. The assembly comprises rectangular cross section tubular members 200, 202 and 204. Tubular member 200 includes rectangular,

(

apertured end plate 210, sidewalls 211-1, 211-2, 211-3 and 211-4 which define a cavity, open end 213, and structural elements 215 and 217 which partition the cavity into sections 214-1, 214-2, 214-3, 214-4. Tubular member 202 5 similarly includes apertured plate end 220 which is secured to open end 213 of tubular member 200, sidewalls 221-1, 221-2, 221-3 and 221-4, and structural members 225 and 227. which partition the cavity of tubular member 202 into sections 224-1 through 224-4. Structural elements 225 and

10227 are skewed with respect to structural members 215 and 217 and apertures 222-1 through 222-4 in plate 220 are offset from apertures 212-1 through 212-4 in plate 210.

Apertured plate end 230 of tubular member 204 is secured to open end 223 of tubular member 202. Structural

15 elements 235 and 237 are attached to plate 230 and to sidewalls 231-1 through 231-4 to partition the cavity of tubular member 204 into sections 234-1 through 234-4. Open end 233 of tubular member 204 is secured to the periphery of transducer 205. The tubular members and the transducer

20 are positioned along a common longitudinal axis through the centers of the tubular members. Each aperture 232-1 through 232-4 in plate 230 communicates a rectangular cross section chamber 238-1 through 238-4 which has apertures in its sidewalls. Chamber 238-1, for example,

25 extends from plate 230 to transducer 205 in section 234-1 and the left end of chamber 238-1 surrounds aperture 232-1. Each sidewall of chamber 238-1 includes an aperture 242 which communicates with section 234-1. Chambers 238-2, 238-3 and 238-4 are similarly arranged in

30 sections 234-2, 234-3 and 234-4, respectively.

The orientation of structural elements 235 and 237 is rotatably offset from structural elements 225 and 227 in plate 220 and the apertures in plate 230 are displaced with respect to the apertures in plate 220. The

35 differently oriented sections of the tandemly arranged cavities in FIG. 7 are operative to inhibit cross mode resonances of the assembled housing and the offset relationship of the apertures in plates 210, 220 and 230

CT

^" .- ^" I

allows each cavity to modify the sound waves passing therethrough independently of the adjacent cavities as aforementioned with respect to FIGS. 1 and 2 and FIG. 4. The equivalent electrical circuit for the 5 arrangement of FIG. 7 is shown in FIG. 8. Resistance 212', inductance 212'', capacitance 214* and conductance 214* ' represent the acoustic characteristics of tubular member 200. Similarly, resistance 222', inductance 222'', capacitance 224' and conductance 224* * represent the

10 acoustic characteristics of tubular member 202. With respect to tubular member 204, resistance 232* and inductance 232* * represent the inertance and loss of the apertures in plate 230, and capacitance 234' and conductance 234* * represent the combined characteristics of

15 the cavity sections of tubular member 204. Inductance 242* represents the equivalent inertance of apertures in the sidewalls of chambers 238-1 through 238-4 and capacitance 241' represents the equivalent acoustic compliance of the chambers. The resonant frequency of chambers 238-1 through

20238-4 may be set to one-half the sampling frequency of the associated digital facility to improve the cutoff characteristics of the acoustic network.

Waveform 901 of FIG. 9 illustrates the frequency response of the arrangement of FIG. 7 for a sampling

25 frequency of 8 kilohertz and a passageway cross section of 42.5 millimeters by 34.0 millimeters where the structural elements of the tubular members are removed. In the absence of the cavity partitioning structural elements, a resonance type response is evident in the region between

304 kilohertz and 6 kilohertz. A double peak response is obtained since two cross mode resonances are significant. Waveform 902 illustrates the frequency response of the assembly of FIG. 7 with the structural elements and chambers placed in the cavities of the tubular members. In

35 accordance with the invention, cross mode resonances are inhibited in waveform 902 and the cutoff at 4 kilohertz, one-half the sampling frequency, is sharper due to the inclusion of resonance chambers 238-1 through 238-4.

The transducers in FIGS. 1, 4 or 7 may be telephone receivers or loudspeaking devices. The function of the receiving transducer assembly is substantially similar to that of the microphone assembly. In a digital communication facility, the receiving transducer assembly replaces the electrical filter normally utilized to eliminate the audio frequency signals above one-half the sampling frequency which are present due to the sampling modulation effect. Where, for example, transducer 205 of FIG. 7 is a loudspeaker of a speakerphone set, sound waves from speaker 205 enter chambers 238-1 through 238-4 and sections 234-1 through 234-4 of tubular member 204. These sound waves are modified in accordance with the acoustic characteristics of tubular member 204 as shown in the electrical equivalent circuit of FIG. 8. The modified sound waves enter cavity sections 224-1 through 224-4 of tubular member 202 via the apertures in plate 230 and exit tubular member 202 via apertures 222-1 through 222-4 in altered form. The altered sound waves from tubular member 202 are further modified in sections 214-1 through 214-4 and in apertures 212-1 through 212-4 of tubular member 200 and the resulting sound waves absent frequency components above one-half the sampling frequency (e.g., caused by sampling modulation) are available from plate 210. The transducer assemblies shown in FIGS. 1 and 4 operate in similar manner when the transducers used are telephone receivers or loudspeaking devices.

While the invention has been described in terms of particular illustrative embodiments thereof, it is to be understood that modifications and alternative constructions may be made by those skilled in the art without departing from the spirit and scope of the invention. For example, the dimensions of the tubular chambers 138-1 through 138-4 in FIG. 4 and 238-1 through 238-4 in FIG. 7 may be made different from each other to extend the range of acoustic network filter characteristics; the angle between cross-member structural element pairs in FIGS. 1, 4 and 7 may be other than 90 degrees to further inhibit

OMPI

cross-mode resonance; or the tubular members may be sectioned into many small divisions to further inhibit cross mode resonance.

OMP /,, WIP