EP0113901A2 - Mode filter - Google Patents

Abstract

A hollow inner conductor (4), expediently having a relatively small diameter, is fitted in a preferably round waveguide (1). The inner conductor (4) has coupling elements, e.g. resonant slots (7, 8) coupling into its interior, at one or more points selected on the suitability of the field, the coupling elements being geometrically reduced with respect to such coupling elements which are also used, or can be used, on the wall of the waveguide - where their function is for external coupling. Having been coupled into the interior of the inner conductor, the wave is converted into a different line wave mode, e.g. the coaxial fundamental wave mode. The wave is then fed out into the space outside the waveguide (1) in this form. The wave mode filter according to the invention can be used as a directional wave filter, e.g. in satellite radio and radar technology, in which the wave mode to be coupled out, for example the H01-wave converted into the coaxial fundamental wave, is made use of to form the antenna deviation information and a further wave mode in the waveguide is used as the useful wave mode which carries the information. <IMAGE>

Description

The invention relates to a wave type switch for the selective coupling out of certain wave types from a waveguide, in particular a round waveguide.

In principle, two methods are known with which a wave-type-selective coupling is achieved. One method is that the type of wave to be coupled out of the waveguide at at least two points on the circumference, e.g. of a circular waveguide, usually not fully broken down into a corresponding number of partial waves, but is only partially coupled in consideration of the signal waves that also couple here and continue to travel in the waveguide. The partial waves of the wave type to be decoupled are then ideally completely recombined - that is, only after the actual coupling process - in their phase position, which only corresponds to this wave type, with an especially necessary microwave network, which is designed in the manner of a bridge circuit. In contrast, those of other types of waves, e.g. the signal waves, inevitably coupled partial waves because of their mutual phase position which is contrary to the wave type to be coupled out, totally reflected by this network. This method, which is called network-selective in the following, is fundamentally associated with a considerable technical outlay because it requires several coupling points and a separate, often relatively complicated network outside the waveguide.

The second method of wave-type-selective coupling is based on the fact that the wave-type to be coupled out of the waveguide is coupled with one of its field strengths Components of such a type and at such a location of the waveguide that none of the wave types to be excluded from the decoupling process has the same field strength component. This method, which is called field-selective in the following, makes the coupling process itself wave-type-selective, independent of the frequency, and avoids any direct influence on other wave types, for example the signal waves traveling on in the waveguide. The field-selective coupling method, if it can be used at all in terms of the field configuration, can usually be implemented with relatively simple means and has the better electrical properties compared to the network-selective method.

The previously known solutions almost always use a mixture of the two methods outlined above. Figure 1 from the essay by G.Mörz "NTZ, Issue 10, Oct. 73, p.441 shows a not exactly simple H20-H _O2 - tracking coupler that works almost only network-selectively. In contrast, those in DE- OS 28 04 132 presented exemplary embodiments of E ₀₁ - H ₁₁ switches completely field-selective.

From the aforementioned DE-OS 28 04 132 an E ₀₁ - H ₁₁ -wave type switch is known, in which the E ₀₁ shaft used as a DF shaft is decoupled from a circular waveguide by means of a coaxial line with an extended inner conductor (longitudinal probe). This longitudinal probe is arranged in the central longitudinal axis of the circular waveguide adjoining the antenna. The capacitive longitudinal probe on the axis of this crossover waveguide is completely decoupled from the H ₁₁ waves of any polarization, for example, and without any selection network.

Basically, E ₀₁ - H ₁₁ crossovers are only able to do this with circular polarization of the signal waves, uie to provide complete information about the storage of the main beam direction of an antenna from the radiation source, for example a satellite.

In addition, the wave-type-selective coupling of the H21 wave and the H ₀₁ wave in the circular waveguide as well as the H ₂₀ , H ₀₂ and E ₁₁ waves in the square waveguide is of direct technical importance, since these wave types are zero when the signal waves are incident axially with 180 ^0- phase jump and their amplitudes for small deposits are proportional to the deposit angle. Furthermore, in certain combinations they are suitable for supplying the complete storage information even in the case of linear polarization of the signal waves. Such combinations are, for example, the E ₀₁ + H21 wave duo, the ^E ₀₁ ⁺ H ₀₁ wave duo and the H _21A + H _21B wave duo in the round waveguide, and the H ₂₀ + H ₀₂ -, the H ₂₀ + ^E ₁₁ - and in the square waveguide the ^H ₀₂ ⁺ E ₁₁ duo. ^H _21A and H _21B waves are understood to mean two mutually orthogonal H ₂₁ waves.

In the general case of linearly polarized signal waves, which are to be expected in practice, double switches are therefore required for these combinations of wave types. The E ₀₁ + H ₀₁ - and the H _21A + H _21B Duo are characterized by the above-mentioned combinations in that they provide the complete storage information regardless of the position of the polarization plane of the signal waves. The double switches for these last two combinations of wave types do not need to be rotated when the polarization plane varies.

The invention has for its object to provide new and simple wave type decoupling options, so that usable for linearly polarized signal waves double switches for each individual wave type to be coupled out of the waveguide and in the waveguide signal waves to be guided essentially meet the requirements of a completely field-selective decoupling, as is already the case according to the arrangement according to DE-OS 28 04 132 for the E ₀₁ -H ₁₁ switches for operation with circular double polarization. In addition, the wave type switch according to the invention should be implemented in such a way that the decoupling of two DF shaft types to be decoupled from one another is at least 20 dB.

According to the invention, which relates to a type switch for selectively coupling out certain types of wave from a waveguide, in particular a circular waveguide, this object is achieved in that a hollow inner conductor is attached in a section of the waveguide, that the inner conductor on one or more , with regard to the field of the respective shaft to be coupled out, has coupling elements which are geometrically suitably selected and which couple into the interior of the hollow inner conductor and which, in terms of their geometrically reduced size, correspond to those coupling elements which are also applied or applicable to the outer side of the waveguide - coupling there reversely to the outside - and that the outcoupled partial waves, each converted into a different line wave type, are recombined starting from the coupling elements and are led out into the space outside the waveguide.

When coupling out certain types of wave (H ₀₁ wave) in the circular waveguide, it is advantageous to arrange the hollow inner conductor coaxially with the central longitudinal axis of the waveguide. In addition, in most cases there are more favorable results if the cross-sectional dimensions of the hollow inner conductor in the Compared to the cross-sectional dimensions of the waveguide are small. In the case of a round waveguide, this means that the diameter of the hollow round inner conductor is small in comparison to the inner diameter D of the round waveguide, for example D / d e5.5.

A central conductor is advantageously provided in the hollow inner conductor on its longitudinal axis, so that the hollow inner conductor and the central conductor running in it result in a coaxial inner line within the hollow conductor. The respective decoupled shaft can be led out via this coaxial inner line after the conversion into the coaxial basic shaft type.

A not only field-selective, but also network-selective mode switch according to the invention can be created if, for the interconnection of several coupling elements, a line network is arranged inside the hollow inner conductor, consisting of combination lines, the combination lines of which originate from a coupling element. These combination lines can then, at their ends facing away from the coupling elements, run geometrically directly into a star point, which is located on the longitudinal axis of the inner conductor, on which the central conductor is optionally arranged and connected to the star point.

To extract the H ₀₁ shaft type from the circular waveguide and to convert it into the coaxial fundamental wave is as. Coupling element in an advantageous manner, a resonance slot running axially in the inner conductor wall is provided, on the one long side of which a conductive sheet metal is fastened to the inner conductor wall, which is attached to it opposite side is mechanically and electrically connected to the central conductor running in the inner conductor. In addition to the one resonance slot, the inner conductor wall can also have one or more axially extending resonance slots which, in their entirety, are arranged uniformly distributed around the circumference of the inner conductor. In this case, the H0 ₁ coupler has a conductive plate attached to the longitudinal side of the slot, which is always on the same side with respect to the slot, on the inner conductor wall directly at the edge of the slot, which is mechanically and electrically connected on the opposite side to the Inner conductor extending central conductor is connected.

For optimal coupling, the coupling elements are arranged at a short distance that results in a maximum coupling, for example λ _H / 4, in front of such a waveguide region in which the cross section of the waveguide tapers or abruptly narrows in such a way that the wave type decoupled by means of the coupling elements no longer is spreadable as a wave.

The hollow inner conductor is expediently held exactly in the center of the waveguide by means of several support struts. One of these support struts can then be used to lead the respective wave type to be coupled out and converted out of the waveguide. Alternatively, the H ₀₁ internal coupler can be integrated with certain polarization switches that have a free center, so that the coaxial line can be led out undisturbed to the rear.

The invention is explained in more detail below with reference to five figures.

• Fig. 1 in einem Diagramm den Verlauf der relativen magnetischen Feldstärke H in Axialrichtung in einer Längsschnittebene durch den Rundhohlleiter für die H₀₁-, H₂₁- und H₁₁-Welle bei gleicher Leistung und einer um 10% über der H₀₁-Grenzfrequenz liegenden Betriebsfrequenz.
• Fig. 2 die Schrägansicht eines H₀₁-Koaxialwellenwandlers,
• Fi^g. 3 eine H₁₁-H₀₁-Wellenweiche, die jedoch für H₁₁-Wellen exakt rotationssymmetrisch ist, in Schrägansicht,
• Fig. 4 eine weitere H₁₁-H₀₁-Wellenweiche mit nur einem einzigen Resonanzschlitz ebenfalls in einer perspektivischen Ansicht,
• Fig. 5 die Schrägansicht einer H₀₁- H₁₁-Wellentypenweiche, die auf dem Prinzip des in Fig.2 dargestellten Koaxialwellenwandlers basiert.

Show it

• Fig. 1 shows in a diagram the course of the relative magnetic field strength H in the axial direction in a longitudinal sectional plane through the circular waveguide for the H ₀₁ -, H ₂₁ - and H ₁₁ -wave at the same power and 10% above the H ₀₁ limit frequency Operating frequency.
• 2 shows the oblique view of an H ₀₁ coaxial wave converter,
• Fi ^g. 3 an H ₁₁ -H ₀₁ -wave switch, which is however exactly rotationally symmetrical for H ₁₁ -waves, in an oblique view,
• 4 shows a further H ₁₁ -H ₀₁ crossover with only a single resonance slot, likewise in a perspective view,
• 5 shows the oblique view of a H ₀₁ - H ₁₁ -wave type switch, which is based on the principle of the coaxial wave converter shown in FIG.

Before going into the figures in detail, the evaluation already indicated with regard to the two coupling options in the waveguide is to be concretized with the following principles. Wherever it is possible to make the coupling field-selective, it is advisable to use it as much as possible even if the decoupling requirements cannot be fully met by this type of selection alone. It is always easier and therefore cheaper, one if possible Large part of the decoupling to be field-selective and only the missing rest to be network-selective.

It is common for the wall of the common waveguide, in which the wave types to be separated run, to be examined first for field-selective coupling options. However, these are not present in the practical interesting example of the H ₁₁ waves of any polarization to be excluded from the coupling, since these waves have every possible field strength component at every point of the hollow tube inner wall. These are, for example, the Er, Hr and H _Z components in the circular waveguide. There is therefore no possibility on the wall of the waveguide of a broadband field-selective coupling compared to H ₁₁ waves.

Certain field-selective effects in cone transitions and also on sudden cross-sectional changes in the common waveguide are not taken into account here because of the strong frequency dependence of the field strength relationships there. Nonetheless, this method of selection can also be used in particular in narrow frequency bands.

In the case of the invention, based on the above negative knowledge, the waveguide wall is abandoned when searching for new field-selective coupling possibilities. The following solution is found. On the waveguide axis, the H ₁₁ waves of any polarization according to FIG. 1 (ordinate for r = 0) to be kept free of coupling have no magnetic longitudinal field strength, but only a magnetic transverse field. In contrast, the H ₀₁ shaft, which is of interest as a DF shaft, has RH in the round waveguide (radius R; -R ≤ r ≤ R) according to Fig. 1 for r = 0, i.e. on the axis, and also a pronounced maximum of the longitudinal magnetic field strength H in the space near the axis. It is therefore possible to couple the H ₀₁ shaft completely on the circular waveguide axis by coupling only with the H _z component, for example with respect to the H ₁₁ waves, that is to say completely or independently of them. This H ₀₁ coupling is also completely field-selective compared to the E ₀₁ wave, because this, like the H ₁₁ waves of any polarization, has no magnetic longitudinal field strength on the waveguide axis. In addition, this H ₀₁ coupling via the H _z component - as can be seen in FIG. 1 - has a high field selectivity compared to the H ₂₁ wave, even with a coupling in near-axis space, because the H _z component of the H ₂₁ - Because of the horizontal tangent on the axis, the distance from the axis increases very slowly starting from zero.

For the interpretation of Fig. 1 it must be added that the relative H _z components of the H ₀₁ , H ₁₁ and H ₂₁ waves are shown on the ordinate at the same power and at a frequency which is 10% above the cut-off frequency of the H ₀₁ wave lies. Therefore, Figure 1 does not yet contain only when H ₀₁ -wave for H _z curve according to Figure 1 still supervening H _z -Überhöhung as a result of its impact against a short-circuit immediately after the Mode filter.

The H _z curves in FIG. 1 show the relative H _z profile in a longitudinal section plane through the circular waveguide RH and its axis. Since this H curve is only rotationally symmetrical in the case of the z H ₀₁ wave, it can also be regarded only as the generatrix of the area with which the spatial H _z distribution of this wave type is described for the H ₀₁ wave. In contrast, the H _z curves for the H ₁₁ and H ₂₁ waves are only shown in such a longitudinal section plane in which H _z becomes maximum
(cosϕ = ¹ for ^H ₁₁ ^{or co} s ^2y = 1 for H ₂₁ ). In a longitudinal section plane perpendicular thereto, the H ₁₁ wave has H _z = 0 (because of cos 90 ° = 0), and the H ₂₁ wave here has an H _z component with the opposite sign compared to FIG. 1 (cos 180 = - 1) with the same amplitude. In a longitudinal section plane generally rotated by the angle ϕ in relation to FIG. 1, the H curve results for the H ₁₁ wave by multiplying the H _z curve in FIG. 1 by cos ϕ and in the H ₂₁ wave by the factor cos 2ϕ.

The invention is now based on the pure H _z coupling found to be opportune on the waveguide axis or in space close to the axis. The practical implementation of this coupling according to the invention consists in the idea of inserting a metallic inner conductor in sections into the previously empty, round waveguide (round waveguide), the diameter d of which is small compared to the outer inner diameter D of the now coaxial waveguide, for example about d <D / 5. Such an inner conductor only slightly changes the field states of the individual H-wave types only in the vicinity of the inner conductor and in no way fundamentally. The field strength components on the inner conductor surface adapt to the known laws, according to which the electric field must be perpendicular to the conductor surface and the magnetic field must run parallel to it. The small field distortions caused thereby, as well as the effects of the inductive and capacitive field displacement by the inner conductor, become smaller the smaller its diameter. An inner conductor with a small diameter d also has no effect on the cut-off frequencies of the most important H-wave types, as it depends on d in the diagram in Fig.4.4 on p.216 in the "Taschenbuch der Hochfrequenztechnik" by Meinke / Grundlach, 1st edition, 1956, Springer publishing house is shown. The line impedance for H ₁₁ waves is reduced by the inner conductor; at a
Inner conductor diameter d = D / 5, the associated reflection is in the easily manageable order of magnitude rnz ≈ 10%: Of course, surface currents flow on the inner conductor in accordance with the law on flooding, each perpendicular to the local, resulting magnetic field strength. According to the article by W.Baier in the magazine "AEÜ", volume 22, issue 4, page 184, the line losses caused by these inner conductor currents, which occur in addition to the outer conductor losses, are at most as large as the outer conductor losses and thus very small if the inner conductor diameters are not too large .

So while the introduction of an inner conductor only has controllable side effects, you gain a variety of new coupling options with very good electrical properties, as shown below.

First of all, it is possible to use one, two or more resonance slots, which are made parallel to the inner conductor axis in the wall of the hollow inner conductor and which are each excited in phase and with equal intensity by one conductor running radially inside the inner conductor, at one of the number of slots to generate a corresponding number of locations on the outer surface of the inner conductor with each other in phase and of equal size, magnetic longitudinal fields. With this field configuration, which does not propagate into the space of the coaxial waveguide with the narrow-band slot resonance, the H ₀₁ wave is excited or decoupled, with a better field selectivity compared to all H ₁₁ and H ₂₁ waves, the thinner the Inner conductor is relative to the outer conductor. The very simple construction of such H ₀₁ -H ₁₁ switches, for example according to FIG. 2, will be explained in detail later.

In addition to this first result of an almost completely field-selective H ₀₁ coupling, the following additional coupling possibilities open up in the coaxial waveguide. This is based on the knowledge that the field strength and current distribution on the inner wall of the outer conductor of a coaxial waveguide - this distribution largely corresponds to that in the hollow conductor for the respective wave type - with the field strength and current distribution on the inner conductor of this coaxial waveguide in the geometric sense is similar. From this, the thesis according to the invention is derived that all coupling mechanisms which have been used on the waveguide wall on the outside so far and which can be used at all can be used for any type of wave, geometrically reduced in size on the inner conductor of a coaxial waveguide. From the inside, they fulfill the same principle function as the outer couplers - as a kind of hollow-world theoretical counterpart to the previous outer couplers.

The progress that can be achieved with the introduction of the inner conductor coupler according to the invention can be summarized as follows. The interconnection of several coupling points is possible with the inner conductor coupler with cable lengths that are more than an order of magnitude shorter than with the previous outer coupler; as the combination of lines running from its origin at the respective coupling point starting directly toward and therefore on shortest paths (λ _0/10 and below), are connected together in a star point as in Figure 3, on the axis of the inner conductor, for example. Such a circuit acts, as will be explained later on using the example, in a very simple manner and with very little frequency dependence on the network. This method of generating network selectivity through an inner network, ie one in the inner conductor located network, thus advantageously stands out from the previously described, previous possibility with an external selection network, ie a network located outside the round or square waveguide, in which the combination lines originating at the respective coupling points are topologically very unfavorable initially radially outwards, So diverge and only have to be brought together again with cumbersome redirection maneuvers, such as after the "NTZ" essay by G.Mörz already mentioned. For these reasons, the outer combination lines must be at least about two wavelengths long.

The very short line lengths in the inner selection network have the further advantage that the individual coupling elements only act on the wave types that are not to be coupled with the sum of their pure reactances, i.e. without the long, short additional external cables at the end. These pure reactances have a significantly lower frequency response than in the known external coupler and can therefore be compensated for in broadband, even if the individual coupling element with partially or only weakly field-selective arrangements for the purpose of complete decoupling of the requested wave type also with the wave type not to be decoupled is strongly coupled. This means that the type of wave to be decoupled is almost completely available at the turnout output and nevertheless the reactance effect exerted by the strong coupling on the type of wave not to be decoupled can be compensated for. It even seems possible that this reactance for the wave type not to be coupled can be used to create other reactances for this wave type, e.g. the post crosses according to DE-OS 28 04 132, to compensate at least partially.

In switches designed according to the invention, the inner conductor must be held in its exactly central position with a holding device. In addition, the DF shaft type or two after conversion, e.g. into the form of the coaxial fundamental wave inside the inner conductor for further use in the space outside the outer conductor of the coaxial waveguide. In addition, an inner conductor holder with coaxial lead-out with good properties e.g. according to the principles and exemplary embodiments shown in DE-OS 28 04 132.

In the following a series of practical application examples of the principles designed above are presented.

2 shows a perspective view of a practically implemented and tested wave type converter which almost completely converts the H ₀₁ wave of the circular waveguide into the fundamental wave of the coaxial line within frequency bands of the relative width of 15%. This H ₀₁ coaxial wave converter consists of a relatively short, outer tubular conductor 1, which can be seen in the foreground of FIG. This outer pipe conductor 1 is initially closed at the rear by a metallic plate 2. The function of this short-circuit plate 2 can be replaced at any time for the H ₀₁ shaft in the circular waveguide by a conical or abrupt transition to a narrower circular waveguide, in which the H ₀₁ field can only be damped aperiodically in the frequency range under consideration, while the H ₁₁ -wave is still well spreadable even in the narrower circular waveguide. With such a narrowing of the cross-section, the H ₀₁ coaxial shaft transition is expanded to an H ₀₁ -H ₁₁ shaft type switch. Such a wave-type switch is shown in FIG. 5, which will be explained later.

An inner conductor 4 is pushed through an axial bore 3 in the short-circuit plate 2 of the arrangement according to FIG. A central conductor 5 is accommodated coaxially inside this inner waveguide 4. This central conductor 5 - hereinafter referred to as coaxial inner conductor 5 - is conductively connected to a transverse plate 6 which branches symmetrically to the left and right as a band-shaped inner conductor. The strip conductor-like sheet 6 runs along the horizontal inner diameter in the inner waveguide 4 and is conductively connected to the inner wall of the inner waveguide 4 on two diametrically opposite longitudinal lines.

If a current flows from the coaxial line coming from the rear right, which is currently directed towards the front, then this current of the coaxial inner conductor 5 branches into two equal partial flows, which continue to flow to the right or left on the transverse plate 6 to the inner wall of the inner waveguide 4. Each of these partial currents flows half according to its origin, for example from the upper semicircle of the inner conductor 5 visible in FIG. 2 on the upper surface of the transverse plate 6, while the other half flows on the lower surface of the plate. Two longitudinal slots 7 and 8 visible in FIG. 2 in the wall of the inner waveguide 4 are thought to be temporarily closed in a conductive manner. If the currents of the transverse plate 6 then meet the inner wall of the inner waveguide 4, the currents flowing on the upper side of the plate 6 from the right and left to the wall are deflected upwards and the currents on the underside of the plate 6 downwards. Immediately after the kink, the currents flow circumferentially upwards and downwards on both sides of the sheet, then bend continuously backwards in the longitudinal direction and are finally distributed in quadrants evenly over the circumference of the inner surface of the inner waveguide 4.

From this current distribution on the inner wall of the inner waveguide 4, there are some practical coupling mechanisms which are realized by introducing coupling openings to the coaxial waveguide at suitable points in the thin wall of the inner waveguide 4 for this purpose.

Fig. 2 shows a coupling arrangement with two longitudinal slots 7 and 8 in the inner shaft wall 4, namely the left slot 7 runs directly above the current-carrying cross plate 6, the right slot 8, however, immediately below. The diameter on which the coupling slots 7 and 8 lie diametrically opposite one another thus has a slight oblique position, but this is insignificant because of the rotational symmetry of the H ₀₁ shaft type to be excited.

If the instantaneous current considered above, which flows on the transverse plate 6 to the longitudinal slots 7 and 8, is now further pursued, it continues as a displacement current flowing through the longitudinal slots 7 and 8, in particular when the longitudinal slots 7 and 8 are in the frequency range under consideration are about half a wavelength long and thus represent resonance slots. The resonance slot 7 or 8 with its displacement current maximum in the middle then directly takes over the line current from the transverse plate 6, which is approximately half as wide as the resonance slots 7 and 8 are long. The possibilities given in the book by Meinke / Grundlach on pages 311 and 312 already quoted for building resonance slots that are much shorter than! o / 2, is pointed out. The displacement current in the left coupling slot 7 of the switch according to FIG. 2 is, for example, for the one generating it, The instantaneous current considered above is directed upwards and therefore, according to the law of flooding, results in a forward magnetic longitudinal field strength on the outer surface of the inner waveguide 4 - in other words in the space of the coaxial waveguide. In the right coupling slot 8, the momentary displacement current is directed downwards and thus also generates a forward longitudinal magnetic field on the outer surface of the inner waveguide 4 which, for reasons of symmetry, has the same amplitude as in slot 7. This H _z field configuration corresponds to H ₀₁ - Wave in the circular waveguide, as follows, inter alia, from FIG. 1.

The technically important question of how the H ₀₁ converter according to FIG. 2 behaves towards H ₁₁ waves depends on the polarization of this H ₁₁ wave. A H _11h wave horizontally polarized in FIG. 2 is already completely decoupled from each resonance slot 7, 8 on its own, because with this polarization the wall currents on the outer surface of the inner waveguide 4 in the area of the resonance slots 7,8 - as well as on that Inner wall of the outermost conductor 1 - only run parallel to the slots 7 and 8. In this converter, in addition to the selection effect explained later by the transverse plate 6, this makes a large contribution to the H _11h decoupling from the coaxial arm. This decoupling is estimated at more than 50 dB.

A vertically polarized H _11v wave in the arrangement according to FIG. ₂ excites the individual resonance _slots 7 and 8, respectively, with its wall currents flowing on the _{inner waveguide} 4, which meet the resonance _slots 7 and 8 transversely, but advantageously in relation to the H ₀₁ wave with a much smaller amplitude. This is shown by the H _z comparison in FIG. 1 with the same power of these two shaft types. There is a high field selectivity, which provides a high H _{11v decoupling} component, which continues to increase as the _{inner waveguide} becomes thinner according to Fig. 1.

The second part of the H _11v decoupling is based on the fact that the H _z components of the H _11v wave are directed opposite one another at the location of the two resonance _slots 7 and 8. The currents passing from the two resonance slots 7 and 8 to the transverse plate 6 and directed towards the tube axis then always have the same direction and, in the case of a symmetrical structure, the same amplitude in accordance with the flow law. Therefore, these radial currents on the transverse plate 6 are always mutually equal, and there is no differential _current that could flow backwards on the coaxial _inner conductor 5, ie the fundamental wave of the coaxial line is completely decoupled from the H _11v wave.

The following result of the electrical field strengths leads to the same result: As explained above, the displacement currents caused by the Hllv wave in the two resonance slots 7 and 8 always have the same spatial direction according to the flooding law. With a time phase that is the same for both coupling elements, this also applies to the corresponding electric field strengths that propagate between the transverse plate 6 and the inner wall of the inner waveguide 4 in the direction of the axis. It is important that of these electric field strengths of the same spatial direction, there is one coming from the left on the upper surface of the transverse plate 6 and the other coming from the right on the lower side of the transverse plate 6. Since, furthermore, the electrical fields on the transverse plate 6 running fundamental wave in one and the same cross-section above and below the transverse plate 6 always in opposite directions, the partial waves of the H _11v wave emanating from the two resonance _slots 7 and 8 are in phase opposition to one another. Your electrical field strengths therefore meet in the Putties always facing each other on both the top and bottom of the cross plate 6. The electric fields of both partial waves therefore always cancel each other out in the vertical longitudinal section plane running through the axis. There is a zero point of the resulting electric field strength, ie a spatially stationary short-circuit plane is created, so that the coaxial fundamental wave cannot be excited in the coaxial line connected here.

The cross plate 6 is a prime example of a very simple, but nevertheless very effective, internal selection network. In interaction with the resonance slots 7 and 8 in the H ₀₁ shaft, it has the effect that its H _z components, which have the same direction in the area of the two resonance slots 7 and 8, produce oppositely directed radial currents on the transverse plate 6. Thus the maximum differential current - that is the scalar sum of these two radial currents - forces an optimal excitation of the coaxial line wave. In contrast, the H _Z components of the H _{11v shaft} , which are oppositely directed at the resonance _slots 7 and 8, always produce radial _currents in the same direction on the transverse plate, which compensate one another on the transverse plate 6 without a coaxial differential current.

The equalizing current flowing on the transverse plate 6 leads to the excitation of an H ₁₁ field inside the inner waveguide 4, ie to a reduced image of the generating H _11v wave in the outer, coaxial waveguide 1. Since the outer waveguide 1 with its H ₀₁ or H11 cutoff frequency determines the operating frequency range, the much thinner inner waveguide 4 provides a strong aperiodic damping for the H ₁₁ field generated in it. In addition to this physical reason for the high decoupling of the H ₁₁ wave from the fundamental wave of the coaxial line, it should be pointed out that the Coaxial line 4.5 still spreading H ₁₁ field remains are totally decoupled from the coaxial fundamental wave.

The compensating _current caused by the H _11v shaft on the transverse plate 6 exerts a certain reactance _effect on this shaft in the central coaxial waveguide 1, 4. According to FIG. 1, this reaction is not strong because of the relatively small coupling H field strengths compared to the H _o1 wave, and it becomes weaker the smaller the diameter of the inner waveguide 4 becomes. Since the reflection of the H ₁₁ passages of these types of wave switches is said to be very small, this reaction will be considered in more detail below.

The physical cause of this reactance _effect can be seen in the fact that the wall currents of the H _11v wave on the _{inner waveguide} 4 generally cannot flow undisturbed via the resonance _slots 7 and 8, but have to overcome a certain reactance there. This reactance is represented by the parallel connection of the pure resonance slot 7 or 8 and the inhomogeneous line system connected to it in the inner waveguide 4, consisting of the right or left half of the transverse plate 6 in FIG. 2 and the inner surface of the right or left half cylinder on the inner wall of the inner waveguide 4.

The two mutually identical resonance slots 7 and 8 each behave in the vicinity of their resonance like a strongly damped parallel resonance circuit. They are therefore high-impedance, so that the wall currents striking the resonance slot 7 or 8 are only approximately the reactance at the input of the above inner conduit system in the inner waveguide 4. This inner line system is formed from the inner surface of the right half cylinder of the inner shaft conductor 4 as the "outer conductor" and the respective half of the transverse plate 6 as the "inner conductor". This line system is short-circuited at the end, specifically in the arrangement of FIG. 2 in the vertical longitudinal section plane running through the axis. This is due to the fact that in this plane the absence of voltage - verified by the decoupling of the coaxial output line - is enforced by the fact that the partial waves of the H _11v waves originating from the two resonance _slots 7 and 8 are equally strong in the middle of the Cross plates 6 - as already explained - meet in phase opposition.

The H _{11v wall} currents on the _inner waveguide 4 are not disturbed when the short circuit prevailing in the central plane is transformed via the inner line system to the location of the resonance slots 7 and 8. This is ereicht at operating frequencies at which half the length of the dash panel 1 _r 6 either is small compared with the wavelength ^- about l _r <λ _o / 10 -, or _r 1 ≈ ₀ .A n / ^2. However, it should be pointed out that with increasing atomic number n, the spatial position of the return locations of this short circuit varies more and more when the frequency changes. There are spatially non-stationary return points of the stationary short circuit in the central plane.

On the other hand, the first H _11v reflection _maximum occurs with increasing frequency when the line length 1 _r = 0.25 λ _or ; because at this frequency and in its environment, the inner line system at the 0rt of the resonance slots 7 and 8 is high-resistance. This influence on the wall currents of the H _11v wave is _alleviated by the fact that, due to the slot length 1 _s > 2 1 _r, the _slot resonance is far below the H _11v reflection maximum, and therefore the slot 7 or 8 is no longer high-impedance here.

It is possible for the first time with the arrangement according to FIG. 2 because of the principle inner waveguide 4 to push the first H _{11v reflection} maximum with 1 _r = 0.25 λ _or so far over the operating frequency range that it no longer interferes. Such a short inner selection network is significantly superior to the previous, outer selection networks with their much longer combination line systems with regard to broadband and especially because of the simplicity of the arrangement.

The H ₀₁ coaxial wave converter according to FIG. 2 can be expanded in the following way to an H ₀₁ - H ₁₁ partial wave crossover, as shown in an oblique view in FIG. 5. The short-circuiting plate 2 in the background of the arrangement according to FIG 5 by a sudden, continuous or stepped transition 16 to a waveguide 17. or waveguide such that only an aperiodically decaying H ₀₁ field can propagate in this continuing waveguide 17, while H ₁₁ - Waves here far enough above their cut-off frequency, and therefore can propagate undamped aperiodically. Such transitions are dimensioned according to known principles. The inner waveguide 4 is either extended to the rear until it is possible at a suitable termination, for example on a polarization switch with a free center, to lead the coaxial line out of the waveguide using the converted H ₀₁ wave. If this is not possible due to the interposition and the pulling through of a shaft inner conductor 4 which complicates elements such as manifolds, the shaft inner conductor 4 is kept as short as possible and is held in its central position with struts extending radially or obliquely to the outer conductor 1. These struts, which are also suitable for leading the inner coaxial line out, can be dimensioned according to the principles set out in DE-OS 28 02 132.

As already mentioned, the resulting H ₀₁ -H ₁₁ crossover is not exactly rotationally symmetrical with respect to any polarized H ₁₁ waves, although the degree of asymmetry is due to differences in reflection (see H _11v - or H _11h Behavior) and in the electrical length of the switch passage paths for orthogonal H ₁₁ waves, because of the relatively very low coupling of the H ₁₁ waves (FIG. 1) is not high. As explained, this asymmetry _{becomes smaller} in the H _01- H ₁₁ switch with two resonance slots 7 and 8 according to FIG. 2, the thinner the inner waveguide 4 is made compared to the outer conductor 1 of the coaxial waveguide.

Absolute rotational symmetry, as is necessary, for example, in operation with circular double polarization, is achieved with the H ₀₁ -H ₁₁ switch outlined in an oblique view in FIG. 3, even with relatively large diameters of the inner waveguide 4. The switch in FIG. 3 works on the same principle of field-selective H coupling as the switch in Fig.2. In addition to the horizontal cross plate 6 of this switch with two resonance slots 7 and 8, in this case consisting of two halves 6 ′ and 6 ″, a vertical cross plate 9 likewise consisting of two halves 9 ′ and 9 ″ is introduced in the arrangement according to FIG. 3, which is contacted in the same way with the inner conductor 5 of the rearward coaxial line as the horizontal transverse plate 6. The newly added, vertical transverse plate 9 feeds two further resonance slots 10 and 11 in the wall of the inner shaft conductor 4. The upper resonance slot 10 is shown in FIG. 3 immediately to the right of the transverse sheet metal section 9 ′ and the lower slot 11 to the left of the transverse sheet metal section 9 ″. The four resonance slots 7, 8, 10 and 11 in FIG. 3 thus lie on the same side of their respective cross from the tube axis sheet section. A current coming from behind on the central coaxial inner conductor 5 is due to the symmetry of the arrangement in four mutually equally strong and in-phase partial currents on the four radially outwardly extending sheet metal sections 6; 6 ", 9 'and 9" split. These currents feed the resonance slots 7, 8, 10 and 11 and generate displacement currents there which are always cyclical and which on the outer surface of the inner waveguide 4 in front of the four resonance slots 7, 8, 10 and 11 in the coaxial waveguide 1.4 each H _z - Excite components of the same strength and direction. This field configuration in turn corresponds to that of the H ₀₁ wave of the round, coaxial waveguide.

For reasons of symmetry, the switch according to FIG. 3 behaves exactly the same with respect to horizontally polarized H ₁₁ h waves as with vertically polarized H _11v waves, namely as described for the switch according to FIG. 2 for H _11v waves . The arrangement according to FIG. 3 thus represents a completely rotationally symmetrical H ₀₁ -H ₁₁ switch with regard to reflection and electrical length for arbitrarily polarized H ₁₁ waves. With a switch designed according to FIG. 3, any depolarization is linear or circular polarized H ₁₁ waves excluded. This also applies if the diameter of the inner waveguide 4 is relatively large, ie if a larger number of displacement current lines meet the inner waveguide 4 and, as a result, a significant coupling of individual resonance slots 7, 8, 10 or 11 with the H ₁₁ waves occurs. Since relatively large diameters of the inner waveguide 4 are to be considered for reasons of manufacturability at higher operating frequencies, for example above 10 GHz, the turnout arrangement according to FIG. 3 is particularly suitable for higher frequencies and at very high frequencies Decoupling requirements between dual circular polarizations.

In addition to the complete rotational symmetry, the switch according to Figure 3 has the following new property. The crossed transverse plates 6 and 9 of this switch act not only as an internal selection network for H ₁₁ waves of any polarization, but also with respect to the H ₂₁ wave. On the outer surface of the inner waveguide 4, the H ₂₁ wave has, at diametrically opposite locations, H _z components of mutually equal strength and direction. These H _Z components, which according to FIG. 1 are very small in comparison to those of the H ₀₁ shaft, would in themselves be capable of a small differential current through the two resonance slots 7, 8 of the one, for example horizontal, transverse plate 6 to generate in the coaxial line. The same applies to the coupler according to FIG. 2, the H ₂₁ selectivity of which is “only” based on the field-selective effect which can be seen, for example, from FIG.

3, however, the H ₂₁ wave with its second pair of H _z components in front of the two resonance slots 10 and 11 of the other, for example, vertical transverse plate 9, produces a differential current which is just as small as the two H components considered first. Since the first and the second H _z component pair in the H ₂₁ wave have the same amplitude but always opposite directions, as is the case with any H21 polarization from the H ₂₁ field image of the coaxial waveguide, this also applies to the corresponding ones Residual currents. These differential currents, which flow in one and the same coaxial line, therefore always cancel each other out.

The H ₂₁ wave is thus decoupled from the coaxial line not only with its bare field selectivity, depending on the diameter of the inner waveguide 4, but also with the network selectivity that the crossed transverse plates 6 and 9 have over the crossover according to FIG H ₂₁ wave is coming. This network selectivity is arbitrarily high with full symmetry of the arrangement.

In the arrangement according to FIG. 3, a fundamental change in function, namely to the H ₂₁ coaxial wave converter, can be brought about with the following simple change. Both resonance slots of one and the same, any transverse plate 6 or 9 are attached to the position shown in Figure 3 on the other side of this transverse plate 9 or 6. A primary current from the coaxial line then generates in the coaxial waveguide 4.1 in front of the resonance slots 7, 8, 10 and 11 H _Z components which have alternating directions in a cyclic order. This H _z configuration now matches the H ₂₁ wave that is excited. In contrast, this alternating H _z component sequence is contrary to the H ₀₁ wave and also to H ₁₁ waves of any polarization. This modified inner selection network now couples the H ₂₁ wave to the coaxial line 5.4 and decouples it from the H ₀₁ wave and from all H ₁₁ waves. However, the network selectivity is reduced here by a negative contribution by "field selectivity", since an H ₀₁ wave running in the coaxial waveguide according to FIG. 1 still excites the individual resonance slots 7, 8, 10 and 11 much more strongly than the H ₂₁ - Wave.

To complete this observation, it is pointed out that the basic shape of the previously described H ₀₁ coaxial crossover is shown in a perspective view in FIG. This design has in a shaft inner conductor 12 only a single resonance slot 13, which is fed by a "half", that is to say only over half a diameter, transverse plate 14 from the coaxial line 5, 12 with the central inner conductor 15. This arrangement works purely field-selective, without any internal selection network. Nevertheless, taking into account the relative H _z amplitudes in FIG. 1, the higher the decoupling of the H ₂₁ wave and all H ₁₁ waves from the coaxial line, the thinner the inner waveguide 12 is to be expected, in particular the H ₁₁ Reflection becomes very small. This simple internal slot coupler is particularly suitable for lower frequencies, for example below 2 GHz.

In the three exemplary embodiments shown in FIGS. 3 to 5, the tubular conductor 1 is not terminated by a metallic short-circuit plate, as in the example according to FIG. 2, but rather passes over to a narrower circular waveguide 17 by means of a conical transition piece 16. In this narrower circular waveguide 17, for example, the H ₀₁ field can only spread aperiodically damped, whereas the H ₁₁ wave can still propagate well there.

Claims (15)

Shaft type switch for the selective decoupling of certain shaft types from a waveguide, in particular a round waveguide, characterized in that a hollow inner conductor (4) is attached in a section of the waveguide (1), that the inner conductor is connected to one or more, with respect to the field of each To be decoupled shaft selected geometrically suitable places coupling elements into the interior of the hollow inner conductor, which geometrically reduced in size correspond to those coupling elements that are also applied or applicable to the outer waveguide wall - although reversely coupling to the outside there - and that each in another Line wave type converted coupled partial waves are recombined starting from the coupling elements and are led out into the space outside the waveguide. 2. Shunt type switch according to claim 1, characterized in that the hollow inner conductor (4) is mounted coaxially to the central longitudinal axis of the waveguide (1). 3. Shaft type switch according to claim 1 or 2, characterized in that the cross-sectional dimensions of the hollow inner conductor (4) are small compared to the cross-sectional dimensions of the waveguide (1), e.g. one in five. 4. Shunt type switch according to one of the preceding claims, characterized in that for interconnecting a plurality of coupling elements an arranged in the interior of the hollow inner conductor (4) from combination lines existing line network is provided which is adequate for the type of wave to be decoupled, the combination lines of which each originate from a coupling element. 5. Wave type switch according to claim 4, characterized in that the combination lines at their ends facing away from the coupling elements run geometrically directly into a star point which is located on the longitudinal axis of the inner conductor. 6. Shunt type switch according to one of the preceding claims, characterized in that a central conductor (5) is provided in the hollow inner conductor (4) on the longitudinal axis thereof, so that the hollow inner conductor and the central conductor running therein a coaxial inner line within the circular waveguide ( 1) result. 7. Shunt type switch according to claim 6, characterized in that an axially extending resonance slot (13) is provided as a coupling element for decoupling the H ₀₁ -wave type from the circular waveguide (1) and for conversion into the coaxial fundamental wave in the inner conductor wall (12) one of the long sides of which is attached to the inside of the inner conductor wall, a conductive sheet (14) which is mechanically and electrically connected on its opposite side to the central conductor (15) running in the inner conductor. 8. wave type switch according to claim 7, characterized in that the inner conductor wall (4) in addition to the one resonance slot (7) has one or more similar axially extending resonance slots (8,10,11), which in their entirety uniformly around the circumference of the Inner conductor (4) is distributed on one longitudinal side of each of the slots (7, 8, 10, 11) on the inside of the inner conductor wall, a conductive sheet (6 ', 6 ", 9', 9") is fastened to the opposite one Side is mechanically and electrically connected to the central conductor (5) running in the inner conductor. 9. Shunt type switch according to claim 8, characterized in that in the case of the H ₀₁ shaft to be coupled along the inner circumference of the inner conductor wall (4), the sheets (6 ', 6 ", 9', 9") always on the same long side of the slots (7,8,10,11) are attached. 10. Shunt type switch according to one of claims 7 to 9, characterized in that the sheets (6 ', 6 ", 9', 9") have a longitudinal extent of approximately half the longitudinal extent of a resonance slot (7,8,10,11) . 11. Shunt type switch according to one of claims 7 to 10, characterized in that the resonance slots (7,8) have a longitudinal extent of about half an operating wavelength. 12. Wave type switch according to one of the preceding claims, characterized in that the coupling elements are arranged at a short distance in front of such a hollow conductor region (16), in which the cross section of the waveguide tapers or abruptly narrows, such that the wave type decoupled by means of the coupling elements is not more than wave is capable of spreading. 13. Wave type switch according to one of the preceding claims, characterized in that the hollow inner conductor (4) is held centrally in the waveguide by means of a plurality of struts engaging on the inside of the waveguide wall. 14. Wave type switch according to claim 9, characterized in that one of the struts is designed such that in each case a wave type to be coupled out and converted can be guided out of the waveguide. 15. Wave type switch according to one of claims 1 to 12, characterized in that the hollow inner conductor (4) is connected at one end to a polarization switch having a free center, and the coaxial line with the converted H ₀₁ wave through the free center, for example is brought out.

Images