Kitazume, Susumu (Tokyo, JA)
Kasuga, Osamu (Tokyo, JA)

05/009926

04/18/1972

02/09/1970

Download PDF 3657670       PDF help

Click for automatic bibliography generation

Nippon Electric Company, Limited (Tokyo, JA)

333/209

333/211

H01P1/16; H01P1/212; H01P1/20; H03H7/10; H01K7/06

333/70,73,73C,73W,95-98,83

3078423	Apparatus for segregating harmonic power in a waveguide system	February 1963	Lewis
3353123	Microwave filter comprising absorbing structures for removing suprious wave energy	November 1967	Met

Saalbach H. K.

Baraff C.

We claim

1. A rectangular waveguide filter having a passband for transmitting electromagnetic waves including a fundamental frequency and attenuating a second harmonic or said fundamental frequency in a fundamental mode TE₁₀₁ keeping resonant frequencies of modes higher than said mode TE₁₀₁ outside of the region where the second harmonic of said fundamental frequency exists comprising:

2. The waveguide filter according to claim 1 in which said capacitive means comprises two adjustable screws having some spaced in mutually parallel relation in said one wide wall in a plane perpendicular to said narrow and wide walls, each screw adjusted to project one end into said cavity; said two screw axes spaced in two positions which include said first-mentioned position and are located said one-half of said one-third waveguide wavelength distance between said two planes including said corresponding axes of said susceptance means and which include each of said two screw axes further located said one-third distance from said inner surface of an adjacent narrow wall and said two screw axes having said one-third distance therebetween.

3. The waveguide filter according to claim 1 in which said capacitive means comprises an adjustable screw disposed in said one wide wall to project one end into said cavity and having an axis in said position which is located said one-half of said one-third waveguide wavelength distance between said two planes including said two susceptance means corresponding axes and which is further located said one-third distance from said inner surface of said adjacent one narrow wall.

4. The waveguide filter according to claim 1 in which said capacitive means comprises two adjustable screws of which one is disposed in said one wide wall and a second in the opposite wide wall; said one and second screws adjusted to project respective one ends thereof into said cavity and having axes in said position which is located said one-half of said one-third waveguide wavelength distance between said two planes including said susceptance means corresponding axes and which is further located said one-third distance from inner surfaces of said respective narrow walls.

5. The waveguide filter according to claim 1 in which said capacitive means comprises two adjustable screws having axes spaced in mutually parallel relation in said one wide wall in a plane perpendicular to said wide walls and parallel to said narrow walls; said two screws adjusted to project corresponding ends into said cavity and to dispose opposing peripheral portions approximately at said position which is located said one-half of said one-third waveguide wavelength distance between said two planes including said susceptance means corresponding axes and which is further located said one-third distance from said inner surface of said adjacent one narrow wall.

6. The waveguide filter according to claim 1 in which said filter means comprises two adjustable screws of which one is disposed in said one wide wall and a second in the opposite wide wall; said one and second screws adjusted to project respective one ends into said cavity and having axes in said position which is located said one-half of said one-third waveguide wavelength distance between said two planes including said two susceptance means corresponding axes and which is further located said one-third distance from said inner surface of said one narrow wall.

7. A rectangular waveguide filter having a passband for transmitting a band of electromagnetic waves including a fundamental frequency and attenuating at least a second harmonic of said fundamental frequency in a fundamental mode TE₁₀₁, comprising:

8. A rectangular waveguide filter having a passband for transmitting a band of electromagnetic waves including a fundamental frequency and attenuating at least a second harmonic of said fundamental frequency in a fundamental mode TE₁₀₁, comprising:

9. A rectangular waveguide filter having a passband for transmitting a band of electromagnetic waves including a fundamental frequency and attenuating at least a second harmonic of said fundamental frequency in a fundamental mode TE₁₀₁, comprising:

10. A rectangular waveguide filter having a passband for transmitting a band of electromagnetic waves including a fundamental frequency and attenuating at least a second harmonic of said fundamental frequency in a fundamental mode TE₁₀₁, comprising:

11. A rectangular waveguide filter having a passband for transmitting a band of electromagnetic waves including a fundamental frequency and attenuating at least a second harmonic of said fundamental frequency, comprising:

This invention relates to a waveguide-type bandpass filter and, more particularly, to a filter of this type capable of rejecting higher harmonic components with the arbitrary selectivity on the fundamental wave component maintained.

In a transmitter-receiver of a microwave communication system, a travelling-wave tube is used as the power amplifier on the transmission side. A travelling-wave tube has, however, inherent nonlinearity, due to which higher harmonics are inevitably generated at the amplification stage. Such harmonic components are not only unnecessary to microwave communication but also undesirable for the system as a whole because it requires excessive power for transmission. Such undesirable components should therefore be removed. A bandpass filter is usually coupled to the output end of the TWT amplifier for this purpose. However, a conventional bandpass filter of the waveguide type is not capable of rejecting the higher harmonic components. It allows the undesirable higher harmonics to pass therethrough together with the fundamental components. To remove the harmonics, a lowpass filter must be employed in addition to the bandpass filter.

It is therefore an object of the present invention to provide a microwave bandpass filter which is capable of rejecting the undesirable higher harmonic components to do away with any additional filter means for the harmonic rejection purpose.

According to this invention, a novel microwave bandpass filter is provided which has sufficiently high rejection characteristic against second harmonic component, which is dominant among the higher harmonic components.

This invention is based on the fact that the higher harmonic components can be substantially suppressed by rejecting the second higher harmonic component, because higher-than-second higher harmonics are very weak and can be neglected. Since the second higher harmonics are in the region where higher transmission modes of microwaves are concentrated, the second higher harmonics can be suppressed by eliminating the higher mode components.

Now, the invention will be described in conjunction with the accompanying drawings, in which:

FIG. 1 schematically shows a conventional bandpass filter;

FIG. 2 is a waveform diagram showing the characteristics of the bandpass filter of FIG. 1;

FIGS. 3a-3f shows various modes possible in a bandpass filter of the rectangular waveguide type;

FIG. 4 shows characteristic curves of the filter to illustrate the principle of this invention;

FIG. 5 schematically shows a bandpass filter embodying this invention;

FIG. 6 shows the characteristic curve of the filter of FIG. 5; and

FIG. 7 through 10 schematically show modifications of the embodiment in FIG. 5.

In FIG. 1, which schematically shows a perspective view of the conventional bandpass filter, susceptance elements 11 and 11', each consisting of three rods disposed in perpendicular relation to the wide plane of a rectangular waveguide 10 for TE ₁₀ mode propagation, are disposed at an interval of one-half of the guide wavelength λg (namely, λg/2) to form a cavity resonator 12. A plurality of resonators 12 and 13 are disposed in series at a spacing of λg/4. Each of the resonators 12 and 13 has tuning screw 14 for attaining the tuned state at each of the resonators 12 and 13.

This bandpass filter has the attenuation vs. frequency characteristic as seen from the curve in FIG. 2 that a bandpass filter consisting of a waveguide having a width a of the wide plane has its selectivity only in the range where the frequency is lower than the cutoff frequency fc ₂₀ for the TE ₂₀ mode. The cutoff frequency fc ₂₀ is approximately equal to c/a; where c denotes the velocity of light. Also, it is seen from the curve that the bandpass filter has its selectivity only in the range where the frequency f is lower than fc ₂₀ . In the region above fc ₂₀ , the characteristics become indefinite, because the existence of possible higher modes disturbs the function of the filter. Generally, the second harmonic 2f ₀ , twice as high as the frequency f ₀ of the passband of the bandpass filter, is higher than fc ₂₀ and included in the region where disturbance is caused.

FIG. 3(a) illustrates field intensity distributions (in absolute value) of the fundamental (or dominant) mode TE ₁₀₁ which is generally used as the desirable frequency of the passband of the filter. FIGS. 3(b) through 3(f) illustrate higher modes TE ₂₀₁ , TE ₁₀₂ , TE ₃₀₁ , TE ₂₀₂ , TE ₃₀₂ and TE ₁₀₃ , respectively, of a bandpass filter consisting of susceptance elements 21 and 21' each comprising three rods disposed in perpendicular relation to the wide plane of a rectangular waveguide 20. Let it be assumed here that the long line of the cross-sectional rectangle of the rectangular waveguide lies in X-axis, the shorter line in Y-axis, and the longitudinal axis in Z-axis. Then, the electric field intensity distribution of the fundamental mode TE ₁₀₁ has a single sinusoidal hump each in X-Y plane and Y-Z plane (extending from the inductive rod 21 to 21'). It should be noted here that the wavelength of the mode under consideration is the guide wavelength. FIG. 3(b) shows a field intensity distribution of higher mode TE ₂₀₁ having a double sinusoidal hump in X-Y plane, and a single hump in Y-Z plane (extending from the inductive rod 21 to 21'). FIG. 3(c) shows a similar distribution of higher mode TE ₁₀₂ having a single sinusoidal hump in the X-Y plane, and a double hump in the Y-Z plane (extending from the inductive post 21 to 21'). Similarly, FIGS. 3(d), (e) and (f) show field intensity distributions of higher modes TE ₃₀₁ , TE ₂₀₂ and TE ₁₀₃ , respectively. To generalize, reference characters "m" and "n" of the notation TE _mon denote the number of the humps of the electric field intensity distribution observed in the X-Y and Y-Z planes.

It is assumed here that the inner width measured in X-direction of the rectangular waveguide 20 is a, and the axial length of the waveguide section or cavity defined by rod arrays 21 and 21' and measured in the Z-axis direction l. Then, the guide wavelength λg of the electromagnetic wave propagated in the waveguide in the TE _mon mode is expressed by

If resonance occurs at the length l of the cavity, then

(n/2)λ g = l (2)

The resonant frequency f at the cavity length is given, from Eqs. (1) and (2), by

When normalized by the use of cutoff frequency f _c (= c/2a) and cutoff wavelength λc (= 2a), Equation (3) is modified into:

Then, from Eq. (4), it follows

because

l/2a = l/λc.

FIG. 4 shows characteristic curves, which are the results of calculation from Equation (5). In FIG. 4, the abscissa represents f/f _c (the resonant frequency for TE _mon mode normalized by the cutoff frequency f _c (= c/2a) for the fundamental mode TE ₁₀₁ ), and the ordinate represents l/λc (the length l of the cavity normalized by the cutoff wavelength λc(= 2a) of the fundamental mode TE ₁₀₁ ). Parameters for these curves are numbers m and n. The legend in the parenthesis along the abscissa is taken to represent the frequency of the passband of the filter and the second harmonic. Curve 31 shows the relationship between the resonant frequency f for the fundamental mode TE ₁₀₁ and the cavity length l normalized by the cutoff frequency and cutoff wavelength for the TE ₁₀₁ mode itself. Curve 32 shows the relationship between the resonant frequency f of a higher mode TE ₂₀₁ and the cavity length l normalized by the cutoff frequency f _c and cutoff wavelength λc of the fundamental mode TE ₁₀₁ . Similarly, curves 33 through 37 show relationships between the resonant frequency f for higher modes TE ₁₀₂ , TE ₃₀₁ , TE ₂₀₂ , TE ₁₀₃ and TE ₃₀₂ , respectively, and the cavity length normalized by the cutoff frequency f _c and cutoff wavelength λc for the fundamental wave TE ₁₀₁ mode.

It will be apparent that the resonant modes having the same number "n" of humps of the electric field intensity distribution observed in the Y-Z plane, for example, TE ₁₀₁ , TE ₂₀₁ and TE ₃₀₁ have similar wavelengths in a waveguide but not in free space.

Generally, a rectangular-waveguide-type bandpass filter is designed to operate with f/f _c value of the desirable (or fundamental) frequency of the passband in the range between 1.4 and 1.8. For example, in WRJ-4-type waveguide for 4,000 MHz band use, f/f _c for the fundamental frequency value is between 1.4 and 1.63. With WRJ-6 type waveguide for 6,000 MHz band use, f/f _c for the fundamental frequency is in the range between 1.58 and 1.71. It is therefore apparent that second harmonic must be rejected in the f/f _c value range between 2.8 and 3.3 or between 3.1 and 3.6. To attain this objective, the resonance curves for the higher modes should never fall within this range. In FIG. 4, the hatched area satisfies this condition. More precisely, the area where l/λc value lies in the range between 0.2 and 0.3 is favorable. The reason for this is as follows: While the l/λc values lie in the region between 0.3 and 0.6, curves 33, 35, and 36 respectively for TE ₁₀₂ , TE ₂₀₂ , and TE ₁₀₃ modes are existent, this is not favorable to elimination of higher modes. Similarly, in the region above 0.6, the resonant frequency for the fundamental mode TE ₁₀₁ corresponding to the fundamental frequency becomes lower than the desirable resonant frequency f _o and very difficult to raise. In the region below 0.2, the resonant frequency for the fundamental frequency is unnecessarily high, and also difficult to lower. Therefore, by way of selecting the cavity length l to fall within this area, the second harmonic component can be rejected. This raises, however, the value of f/f _c of the fundamental frequency to a value ranging from 1.9 to 2.7, as shown by the curve 31. To restore this to the range between 1.4 and 1.8, a capacitive element must be inserted into the cavity of FIG. 3(a) and thus to reduce the resonant frequency for the fundamental mode TE ₁₀₁ . Briefly, in the conventional filter, the cavity length l is taken in the range between 0.4 and 0.5 in the 1/c (or 1.4 to 1.8 in f/f _c ). Therefore, the higher modes appear within the frequency region above f _c20 where disturbance is caused and where the second harmonic 2fo is included. In this invention, therefore, in order to prevent the resonance of the second harmonic 2fo, the length l is selected in the range from 0.2 to 0.3 in 1/c, wherein the higher modes on the second harmonic 2fo do not exist and the resonant frequency for the desirable (or fundamental) frequency is reduced. In this case, the resonant frequencies of the higher modes TE ₂₀₁ , TE ₁₀₂ , TE ₃₀₁ , TE ₂₀₂ and TE ₁₀₃ must be arranged so as not to allow f/f _c to come in the range between 2.4 and 3.6, because this region belongs to the undesirable second higher harmonics, which could possibly be resonant to the higher modes. The method for attaining this is as follows:

As indicated by the curve 32 in FIG. 4, it is sufficient for the higher mode TE ₂₀₁ to reduce its resonant frequency or to keep it unchanged. As shown in FIG. 3(b), the position at which the field of TE ₂₀₁ mode is minimum on the center line E-E' on the major plane of the rectangular waveguide 20. This means that the resonant frequency of TE ₂₀₁ mode can be reduced by inserting a capacitive rod at a position except for on the line E-E'. In the TE ₂₀₁ case, the cutoff frequency f _c20 normalized by the cutoff frequency f _o for TE ₁₀₁ mode, f _c20 /f _c , is equal to 2, which is well above the frequency region for TE ₁₀₁ mode ranging from 1.4 and 1.8. In other words, as for TE ₂₀₁ mode, the capacitive rod may be disposed anywhere. Also, since the resonant frequencies for higher modes TE ₁₀₃ , TE ₄₀₁ , TE ₃₀₂ are high enough, there is no problem in rejecting the second harmonic. Therefore, consideration must be taken only for the modes TE ₁₀₂ , TE ₃₀₁ and TE ₂₀₂ . In order to prevent the resonant frequencies for the three modes TE ₁₀₂ , TE ₃₀₁ and TE ₂₀₂ from coming down to reach a certain specific band as a result of insertion of capacitive rod into the cavity, this capacitive rod must be disposed as such a point at which the field of each the higher modes TE ₁₀₂ , TE ₃₀₁ and TE ₂₀₂ is minimum. Also, in order to lower the resonant frequency of the fundamental wave TE ₁₀₁ , the capacitive rod must be disposed at such a point at which the field of the fundamental mode TE ₁₀₁ is maximum as the center position of the wide plane of the waveguide which forms the resonator with the susceptance elements. For the TE ₁₀₂ mode, the capacitive rod should be disposed on the center line A-A' of the cavity at which the field of the same mode is minimum as shown in FIG. 3(c). For the TE ₃₀₁ mode, the capacitive rods should be positioned on the trisectional lines B-B' and C-C' on the major plane along the longitudinal axis at which the field of the mode is minimum as shown in FIG. 3(d). For the TE ₂₀₂ mode similarly, the element is on the center line A-A' of the cavity and also the center line E-E' on the major plane of the rectangular waveguide, at the two lines the field is minimum as shown in FIG. 3(e). The positions common to the conditions for these modes are D and D' at which the center line A-A' of the cavity are in crossed relation with the trisectional lines B-B' and C-C' as shown in FIG. 3(d). As described above, the TE ₂₀₁ mode has no problem in rejecting the second harmonic component. In order to reject the TE ₂₀₁ mode, it is necessary to make the arrangement of the bandpass filter symmetrical because the field intensity distribution on one side is in opposite phase with that on the other side with respect to the center line E-E' on the major plane, and because a mode having opposite phase does not occur within a symmetrical waveguide.

Embodiments of the invention will be further described referring to FIGS. 5 and 6. In order to prevent the resonant frequencies for higher modes other than TE ₂₀₁ from falling in the region below the frequency twice as high as the resonant frequency for the fundamental wave, the length l of the cavity of the rectangular waveguide section is made equal to 1/3λg, in contrast to the corresponding length 1/2λg of the conventional bandpass filter and adjustable capacitive elements (screws) 41 and 41' are installed at two points, respectively, at which the trisectional lines 42 and 42' on the major plane of the rectangular waveguide 40 intersect with the bisector 43 of the interval l between the induction rods 44 and 44'. As will be understood from FIGS. 3(c), (d) and (e), and the description thereof, these points correspond to the points where the field intensity is minimum with respect to the higher modes TE ₁₀₂ , TE ₃₀₁ and TE ₂₀₂ , and the same is substantially maximum with respect to the fundamental mode TE ₁₀₁ . By the use of the adjustable capacitive screws 41 and 41', a desired passband can be attained for the fundamental mode TE ₁₀₁ , keeping the resonant frequencies for higher modes outside of the area wherein the second harmonic of the fundamental frequency exists. Also, the TE ₂₀₁ mode is rejected due to the symmetry of the filter.

FIG. 6 shows the attenuation vs. frequency characteristics of the second-harmonic-rejecting bandpass filter of FIG. 5. As is apparent from the characteristic curve, this filter is capable of rejecting the higher modes in the frequency region twice as high as the resonant frequencies for fundamental mode. More specifically, this filter rejects the higher modes in the region where f/f _c ranges from 2.8 to 3.6, while the f/f _c value for the passband for fundamental wave ranges from 1.4 to 1.8. Thus, the second harmonic wave component is eliminated.

FIGS. 7 through 9 show modifications of the embodiment in FIG. 5. These modifications are based on the fact that, as described above, the higher mode TE ₂₀₁ may be left out of consideration if the second harmonic is to be rejected and, therefore, the structure of the bandpass filter need not be symmetrical.

FIG. 7 shows a modification of the embodiment in FIG. 5 wherein only one of the adjustable capacitive elements (screws) of FIG. 5 is used. FIG. 7 embodies only the adjustable capacitive element 41 disposed at the intersection of the trisection line 42 and the bisector 43 in the manner of FIG. 5, the adjustable capacitive 41' in FIG. 5 being omitted in FIG. 7. The essential dimensions in FIG. 7 being the same as corresponding dimensions in FIG. 5.

FIG. 8 shows another modification wherein one of the adjustable capacitive elements (screw) is installed at the trisectional in the opposing major surface of the waveguide. FIG. 8 embodies the adjustable capacitive element 41 disposed at the intersection of the trisectional line 42 and the bisector 43 in one waveguide wide side in the manner of FIG. 5 but the adjustable capacitive element 41' is disposed at the intersection of the trisectional line 42' and the bisector 43 in the wide side of the waveguide opposite to that embodying the adjustable capacitive element 41. The essential dimensions in FIG. 7 being the same as corresponding dimensions in FIG. 5.

FIG. 9 shows still another modification wherein two capacitive elements 41 and 41' are spaced on trisectional line 42 on opposite sides of bisector 43 in the vicinity of the position corresponding to that of the capacitive element 41 of FIG. 7. The essential dimensions in FIG. 9 being identical with corresponding dimensions in FIG. 5. Likewise, the arrangements of FIGS. 5 and 8 may be modified by replacing the single capacitive element with a plurality of capacitive elements disposed at around the positions as in FIG. 5 or 8.

FIG. 10 shows an arrangement wherein the capacitive elements 41 and 41' are oppositely installed on the opposite major planes of the waveguide. Accordingly, in FIG. 10, element 41 is installed at the intersection of trisectional line 42 and bisector 43 in the upper major plane of the waveguide while element 41' is installed at the intersection of the trisectional line 42 and bisector 43 in the lower major plane of the waveguide, whereby the elements 41 and 41' are oppositely disposed in opposite major planes of the waveguide. The essential dimensions in FIG. 10 are the same as corresponding dimensions in FIG. 5. The similar arrangement may be made in connection with FIGS. 5 and 7.

In the embodiment and modifications, single stage bandpass filters have been described. Generally, the higher harmonics-rejecting bandpass filter consists of a plurality of stages. Needless to say, the invention can be applied to such multi-stage bandpass filters. The invention is applicable also to the 1/4-wavelength-coupling-type and direct-coupling-type higher-harmonics rejecting bandpass filters.

The rectangular waveguide employed in the above embodiment and modifications may be replaced by a circular or elliptic waveguide. Also, the number of the susceptance rods employed in the embodiment to define each stage of the filter may not necessarily be three. It may be two, four or any other arbitrary number. Furthermore, these susceptance rods may be of window shape or any other shape.

In the embodiment, two adjustable screws are employed as the variable capacitive element at symmetrical points on the major plane or planes. However, the number of the screws may be chosen arbitrarily. The positions of the capacitive elements may not necessarily be symmetrical.

Also, it will be apparent to the engineers in this technical field that the principle of the present invention is applicable to rejection of higher-than-second higher harmonics.

While the invention has been shown schematically and described in detail with reference to particular embodiments and modifications, it will be clearly understood that the general principles of this invention may be applied to those skilled in the art to other structures of the microwave bandpass filter without departing from the spirit of the invention.