DE977019C - Coupling arrangement for shaft transmission systems - Google Patents

Description

(WiGBI. P. 175)

ISSUED DECEMBER 23, 1964

P 30093 IXdI'21 a ⁴ D

has been named as the inventor

The invention relates to a coupling arrangement for coupling four transmission circuits A, B, C, D of a wave transmission network. It is known to couple four transmission circuits as branches to a closed transmission loop, the electrical length of which is an odd multiple of a half-wavelength at the operating frequency, in such a way that the wave transmission between each circuit of a first pair comprising circuits A and C and each circuit of a second pair comprising circles B and D can take place, but that the transmission of waves between circles A and C or B and D of the same pair is largely prevented.

In the known coupling arrangement, all connections are designed as parallel connections. This results in certain restrictions in use; in particular can use the known arrangement as a waveguide double, compensator only in such a way that in one side of the loop a conversion or conductor reversal is made to on the opposite connection a cancellation of the propagating in the two sides of the loop Realize wave energy.

409 765/5

The invention aims to perfect a coupling arrangement of the type indicated, and in particular in the implementation of double compensation, the dependence on conversion or avoid interchanging measures.

The special feature of the invention is that using either a parallel or a series connection for each branch at least one of the branches by means of a series connection

ίο the closed loop is connected that the along the loop existing electrical distance between two each connected next to each other Branch circles approximately zero or a whole multiple (o, i, 2, 3, 4) of the quarter wavelength at operating frequency and is the distance along the loop between the two circles of each pair approximated to zero or an even multiple (0, 2, 4, 6 ...) of the quarter wavelength is rated at operating frequency if the connections to the loop are of the same type, i. H. both Series or parallel connections are that the distance between the branches of a branch along the loop Pair when using a series connection for one branch and a parallel connection for the other branch is approximately an odd multiple of a quarter wavelength and that at most three of the four branches are connected at any given point on the loop.

The connection of a branch line with a main line (loop part) becomes a series connection called when, when power is supplied from the branch line to the main line, the phase of the waves; which advance in the main line away from the junction in one direction, by 180 ° deviates from the phase of the waves moving away from the junction in the other, i.e. H. opposite Direction progress, the phase in principle simultaneously on both sides of the junction is measured at points equidistant from the center of the latter.

The connection of a branch line with a main line (loop part) becomes a parallel connection called when the phase of the waves, when power is supplied from the branch line to the main line, which proceed in the main line away from the junction in one direction with the Phase of the waves coincides, which is away from the junction in the other, d. H. opposite Direction progress, the phase in principle simultaneously on both sides of the junction is measured at points equidistant from the center of the latter.

In a further embodiment of the invention it is proposed that that the transmission loop has the same wave resistance over its entire length and that each circuit of the one pair has an impedance equal to that at its loop junction Characteristic impedance of the loop circle, while the impedance of the two circles of the other Pair is dimensioned at their loop connection points so that the sum of their impedances through the loop circle seen over each of the first-mentioned circles is also equal to the wave resistance of the loop circle is. According to the invention, the loop circle can be designed in a very versatile manner Experienced. In particular, there is the possibility of designing the loop circle as a coaxial line or as a shielded two-wire line; the loop circle can also consist of a Ring-shaped waveguides for polarized electromagnetic waves exist, their electrical Vector runs parallel to the loop plane. There is also the option of using the Loop circle made up of two or more networks in series with concentrated circle constants each of which is essentially a quarter-wavelength transmission line is equivalent at the predetermined operating frequency.

Further characteristics of the invention emerge from the following description in connection with the drawings, in which FIGS. 1 to 10 serve only for general explanation, but do not yet illustrate the actual invention. Fig. Ι shows a simplified diagram;
2 to 10 show partial cross-sections and perspective views of various arrangements which are used to connect waveguides to one another or to connect waveguides with coaxial lines, so that series parallel connections are formed, as are used in the arrangements according to the invention;

11 shows a perspective view of a type of coupling arrangement within the meaning of the invention;

Figs. 12, 13 and 14 are diagrams showing the Illustrate the operation of the shaft coupling arrangement according to FIG. 11;

15 to 29 show perspective views or cross-sectional representations of other coupling elements or compensation arrangements within the meaning of the invention, with waveguides and coaxial lines in combination as well as various combinations of series or parallel connections of the Branch lines on the coupling loop apply 10; find dung;

30 to 34 show other coupling or compensation arrangements within the meaning of the invention, wherein the coupling loop or the coupling ring consists of coaxial cable and the an- in coupled branch lines are coaxial cables or shielded double lines or parallel wires represent;

35 to 39 illustrate further coupling or combination arrangements as defined in ii | of the invention, wherein the coupling ring or the coupling loop and the branch lines are made of shielded There are double lines;

40 to 44 show other coupling or compensation arrangements according to the invention, wherein the coupling ring or the coupling loop is made up of circles with lumped constants is composed;

45 to 48 illustrate the application of the coupling or compensation arrangements iaj according to the invention in double message systems

Men with communications by wire or wireless, in a location system and in an intercom amplifier;

49 shows in schematic form how a plurality of coupling or compensation arrangements according to the invention are combined to form multiple intercom equalizations.

The scope of the principles which the ίο mode of operation of the device according to the invention so general that other forms of wave propagation are included and that Restrictions in the use of the device mainly due to the prerequisites the practical construction are given. For the sake of simplicity, the following, ins single walking description on formations relating to AC power transmission and some of the most important possible use cases are limited.

Since all the devices of the invention work with the aid of wave compensation, and there the devices mainly useful for double messaging systems where they are concurrent They will allow transmission of messages in two directions at the same frequency in the following description, for the sake of convenience, with the general expression "double compensator" designated. However, this is intended to be used in the description and in some claims Experience a broad interpretation and not be limited to double systems.

The embodiments of the invention described below, in which hollow metal waveguides, coaxial Cables, shielded wire lines or combinations of such conductors are used, are particularly suitable for systems that transmit waves at ultra-high frequencies.

The designs in which circles with lumped constants, e.g. B. Coil-capacitor networks Find use are particularly suitable for systems which waves of relatively transmitted at a low frequency.

The device according to the invention is to be described in general terms using the simple diagram in FIG. 1. In this figure, the device is represented by a box, and it is irrelevant for the general description what is in the box. However, the content should be such as to achieve the results described below. Four "lines", labeled A, B, C, and D, extend from the box. Such a line can consist of a waveguide, a coaxial cable, a shielded pair of conductors, a pair of wires and another conductor that is suitable for the frequency present. If a generator is connected to line A and the compensation has been carried out and suitable loads are connected to the other three lines, the power output by the generator is evenly distributed between the loads at B and D , while the load at C has no power is recorded. If the generator is connected at B , the power is evenly distributed between the appropriate loads at A and C , while no power goes to the load at D.

When using such a device, a signal transmitter is connected to line ^, an artificial load is connected to line D , a signal receiver is connected to line C, and a communication line is connected to line B. The receiver is not affected by the power that goes from the transmitter to the communication line and the artificial load; however, it responds to the performance coming from the message line.

As the use of hollow metal waveguides in certain embodiments of the invention requires special connections and since this mode of operation differs considerably from the connections in the case of coaxial cables, parallel wire lines or electrical networks appears It is desirable at this point to learn some basics of transmission with waveguides and the construction of waveguides.

Basics of waveguide technology

Of the infinite number of possible wave types that propagate in hollow metal waveguides only the fundamental wave should be considered. The fundamental wave is the wave with the lowest critical frequency. The reason that speaks most for using this wave is is that the fundamental wave can propagate when the frequency used is in the middle between the critical frequency of the fundamental wave and the critical frequency of the wave with the next higher order. This avoids an unfavorable mix of wave types, which can arise from irregularities and imperfections in the structure of the waveguide. For this reason it is favorable that the restriction to the fundamental wave should be the simplest and the most powerful form of a double compensator for waveguides. However, none seem basic causes, which is why other wave types are not used could.

There are waveguides with circular and rectangular cross-sections. The rectangular waveguide are superior for many purposes because the fundamental wave only propagates with one polarization, when one side of the cross section is sufficiently small. That is, reproduction takes place so instead that the electric field strength is parallel to the small side. In the executions According to the invention, therefore, rectangular waveguides are preferred.

It is also useful for understanding the operation of double compensators for waveguides Knowledge of the basics of branching in waveguides is required. The general case such a junction is very complicated, but there are certain special features discussed below.

specific cases that can be examined qualitatively. Let it be B. Fig. 2 viewed. This figure shows a straight piece of rectangular waveguide to which another rectangular waveguide is connected at right angles, namely in the electrical plane, ie in the plane which is parallel to the lines of force of the electric field strength of both conductors. In Fig. 2 the lines of force of the electric field strength are drawn according to the Huygens principle in order to qualitatively describe the process in which the fundamental wave moves from the vertical arm towards the connection point of the two conductors. Although a certain reflection can occur at the junction, it can be seen that the "transmitted power" is evenly distributed between the two arms lying in a line and that the waves at AÄ are phase-shifted by 180 ° from the waves at BB 'when the Cross-sections AA ' and BB' are equidistant from the junction.

Fig. 3 shows the same junction as Fig. 2. Here, the solid lines within the conductors represent the electrical field strength of the waves arriving from the left and the dashed lines the electrical field strength of the waves arriving from the right in the two parts lying in a line. It can be seen that the vertical arm receives waves which are phase shifted by 180 ° when the two wave trains at AA ' and BB' are in phase. If the amplitudes of the two original wave trains are the same, the waves in the side arm cancel each other out and the branch consumes no power. Such wave trains of the same amplitude, which progress in the opposite direction, naturally generate standing waves with a tension bulge at the connection point. In other words, a vertical arm that is "introduced" into the electrical plane at the voltage maximum of a standing wave does not consume any power.

4 shows a junction of waveguides in the magnetic plane, ie in the plane which is parallel to the lines of force of the magnetic field strength and therefore perpendicular to the lines of force of the electric field strength, which is denoted by small circles. The lines of electric field strength are shown for waves that progress from the side arm towards the junction. Since the geometric arrangement does not work in such a way that the polarity of the lines of force is reversed, the waves in the cross-section ^^ 'are in phase with those in the cross-section BB'. The side arm will receive the waves in the case of waves with the same amplitudes arriving from the left and right and which are in phase at the cross sections AA ' and BB' , since all waves are in phase. So it will consume a maximum of power. In other words: the side arm consumes a maximum of power when it is introduced where there is otherwise a maximum voltage of the standing wave. For the above statements about branches in the electrical and magnetic planes to be correct, it is not necessary that the two arms of the waveguide be in line. 6g It is necessary, however, that the branch be symmetrical with respect to the side arms. Fig. 5 shows such a symmetrical junction in which the arms of the waveguide are not in a line.

Because of the analogy with connections with the usual transmission lines, the device can 2 as a series connection of the side arm with the "main waveguide" and the connection 4 are viewed as a parallel connection because of the prevailing Phase relationships. In the following description, the devices of Fig. 2 and after FIG. 4 denotes the generalization according to FIG. 5 with a series connection or with a parallel connection.

Fig. 6 shows the device which is suitable for a branch in a connection of waveguides with a coaxial cable. The power supplied by the coaxial line, which is shown schematically in the middle of the figure, spreads to the right and left, the phase shift in the cross-sections AÄ and BB being ' o °. Conversely, in the case of waves of equal amplitude traveling to the right and left in the waveguide, the coaxial cable will absorb maximum power if the waves moving to the right in cross section AÄ are in phase with the waves moving in the opposite direction in cross section BB ' are. A comparison of this process with the process in the device according to FIG. 4 shows that the device according to FIG. 6, at least as far as the phase position is concerned, can be referred to as a parallel connection of a coaxial line with a waveguide. There are several other ways to connect a coaxial cable to a waveguide; z. B. needs the inner coaxial conductor only partially protrude into the waveguide, or it can be closed within the waveguide by a suitable antenna, the other end of which has a return path to the outer coaxial conductor. The connection can even be made in the magnetic plane provided that the inner coaxial conductor is terminated by means capable of absorbing the power of the waveguide. It must therefore be emphasized that there are also other methods of making the connection between coaxial cables and waveguides which are more suitable for special applications, but the arrangement shown in FIG. 6 is mostly used for the schematic illustration in this invention. ia <

The most important findings about the special cases of junctions described above can be summarized as follows:

ι. A generator in a vertical or symmetrical Sidearm of a waveguide gives 12; to the other two arms waves off that

have a phase shift of 180 °, when the side arm is connected in series with the other two arms. One Such a series connection is a junction in the electrical plane, as shown in FIG. 2 is shown (or in more general form in Fig. S).

2. A generator in a vertical or symmetrical The side arm of a waveguide emits waves to the other two arms that are in phase when the side arm is connected in parallel connected to the other two arms. Such a parallel connection is a branch in the magnetic plane, as shown in Fig. 4 (or more generally in Fig. 5), or a connection with a coaxial line as shown in FIG.

3. A symmetrical series connection with a waveguide with a standing wave does not consume any power if the connection is made at a voltage maximum is. It consumes maximum performance when the connection is made there is where there is otherwise a tension node.

4. A symmetrical parallel connection with a waveguide with a standing wave does not consume any power if the connection is made at a voltage node is. It consumes maximum power when the connection is added there is where there is otherwise a stress maximum.

As will be stated later, some forms of double composers are in accordance with the invention necessary to attach two or three connectors to a waveguide, namely to the same place. Concerning the superposition of two connections interested in double expansion joints only the case that there is a series connection and a parallel connection. Fig. 7 illustrates one way of making such a connection, using two waveguides with one Main waveguides are connected. The series connection takes place in the electrical planes while the parallel connection is made in the magnetic plane.

In Fig. 7, the rectangular waveguides 1 and 2 are in the same place with the rectangular one Main waveguide 3 connected in the form of two branches. The branch waveguide 1 is connected to the main waveguide 3 in the electrical plane, d. H. in the plane parallel to the lines of force of the electric field strength of both waveguides. This connection can, as at Hand of Fig. 2 was explained above, as an electrical series connection to the main waveguide be considered. The branch waveguide 2 is in the magnetic with the main waveguide Level connected, d. H. in the plane parallel to the lines of force of the magnetic field strength lies. This connection can, as was explained above with reference to FIG. 4, as an electrical connection Parallel connection with the main waveguide. Fig. 8 schematically shows another Possibility for such a connection. A coaxial cable is used here for junction 2, in order to obtain an electrical parallel connection with the main conductor 3. Because of the presence of the waveguide row junction 1, it is not possible to use the device of FIG. 6 in the form shown. Therefore, another of the above-mentioned designs is used. In the embodiment shown in Fig. 8 it is provided for voting that the length of the inner coaxial conductor 4 transversely to the waveguide 3 is variable.

The various general forms of basic interest with three superimposed connections on a main waveguide are summarized in the following table:

Type of connection

line line Parallel Parallel a
b
C.
d
e Waveguide
Waveguide
Waveguide
Waveguide
Waveguide Waveguide
Waveguide Waveguide
Waveguide
Coaxial cable
Waveguide
Coaxial cable Waveguide
Coaxial cable
Coaxial cable

The arrangements α and b can easily be seen from FIG. 7; Fig. 8 applies to d . Possible arrangements for c and e are shown in Figs.

It is appropriate to start this discussion of waveguide fundamentals with a treatment of wave impedance of waveguides to decide. There are several definitions according to which the wave resistance can be calculated from waveguides. These different calculations lead to iao Expressions which have the same functional dependence on the dimensions of the conductor and on the Show frequency; however, they differ in their numerical constants. The fact that there is an ambiguity with regard to the absolute value of the wave resistance, normal

409 765/5

sometimes irrelevant in practical calculations. That is to say that it is common only concerns relative quantities and that for such cases the unambiguous use of an impedance expression, which is based on some definition will lead to the same results as they are obtained with reference to another definition.

The wave resistance Z ₀ of waveguides with

ίο circular and rectangular cross-section can be easily calculated. For the dominant wave in circular conductors, Z ₀ changes from infinity at the. critical frequency down to a value of a few hundred ohms for frequencies that are remote from the critical frequency. For example, in a calculation based on transmitted energy and shunt current _{, Z 0 approaches 353 ohms as the frequency is increased.} In order to obtain a wide range of impedance values it becomes necessary to approach the critical frequency so closely that experimental difficulties can often arise. For the dominant wave in the rectangular conductor, _{however, Z 0} not only has the same dependence on the frequency, but also a linear dependence on the dimensions parallel to the electric field. A wide range of impedance values can thus be obtained by varying this dimension. In certain cases, this fact leads to a preference for rectangular over circular conductors. For the rectangular conductor, however, it is advisable to limit the dimension in the electrical plane to such values that the dominant wave for the polarization perpendicular to the chosen direction cannot be maintained.

One type of double waveguide compensator

A step-by-step explanation of the construction and operation of a particular form of Double waveguide compensators according to the invention are given in connection with FIG. In Fig. 12 is a closed ring or loop-shaped waveguide drawn, the middle Circumference of one and a half wavelength and at which a straight waveguide is symmetrical is connected, d. that is, the axis of the straight conductor runs through the center of the ring.

The plane of the ring is the electrical plane. Fig. 12 thus shows a cross section in the electrical plane, and the electrical vector for the entire waveguide lies in the plane of the paper.
Assume that a generator is attached to the sidearm waveguide to send fundamental waves toward the ring. In FIG. 12, vectors are shown which represent the electrical field strength existing for a moment with reference to time. The solid and dashed vectors are assigned to the waves which have progressed to the left or to the right at the junction between the straight waveguide and the ring. The effect is similar to that which has been discussed in connection with the branch arrangement 6g according to FIG. 2, especially since the waves diverging to the left and right at the junction are shifted in phase by 180 °. Since the ring comprises an odd number of half-wavelengths, the waves which have traversed the ring reach the junction in phase with those waves which enter the ring. Since the generator does not deliver any other power to the ring than the negligible ohmic losses, in the steady state there is an essentially pure standing wave in the entire ring, with a voltage maximum at the connection point. The behavior over time can easily be derived from the instantaneous values according to FIG. 12 by letting the solid and dashed vectors oscillate between oppositely polarized limit values of the same magnitude. In addition to the voltage maximum at the connection point, there are two further maxima and three voltage nodes, the distances being equal to a quarter wavelength, as shown.

If openings are cut in the ring at points 2 and 4 and the waveguide is symmetrical are connected to the ring at these points, it follows from the principles discussed above, that these series connections receive maximum power. If in these branch waveguides If the same loads are applied, they will also absorb the same power. The amplitude of the waves passing point 4 towards point 3, will be equal to the amplitude of those waves which pass point 2 in the direction of point 3 go, as the same services are drawn from the waves as they pass. this part of the ring between points 2 and 4 will therefore still contain only one standing wave, with a voltage maximum at 10; Position 3. One attached to point 3 of the ring Series connection will not receive any power for this reason.

Fig. 11 shows a perspective view of such a modified embodiment, where the n < four straight waveguide branches by means of series connection with the waveguide ring at the points i, 2, 3 and 4 are connected.

It will now be investigated what happens if a generator in the waveguide iij arm is attached to point 2 of FIG. As can be seen, this case can be derived from the above Derive the case by rotating all parts by 60 ° in FIG. 12 and only the Except numbers. In this way the 12c interference of the fields is used to produce a voltage maximum at points 2, 4 and 6 and to create tension nodes at the points i, 3 and 5 of the ring contribute. If now the same loads in the branch waveguides 1 i2 £ and 3 are attached, the same are obtained from '

the generator, which is assigned to the waveguide branch 2, the same power, while the load in the waveguide branch 4 does not consume any power, since this branch is at a voltage maximum connected to a pure standing wave.

The arrangement of Figure 11 has been shown to be a suitable construction for a double waveguide compensator as far as compensation is concerned. It will now be shown that a set of impedances can be selected so that impedance matching is achieved, regardless of whether the compensator is operated by the waveguide branch 1 or by the waveguide branch 2. Checking the impedance values according to FIG. 11 results in the following: If the wave resistance of the waveguide forming the ring is _{chosen to be Z 0} , then branches 1 and 3 consist of waveguides with the wave resistance 2 Z ₀ , while

ao the branches 2 and 4 consist of the waveguides with a wave impedance Z ₀ . It is also assumed that the waveguide branches 2, 3 and 4 are terminated with their wave resistances and that a generator of the impedance 2 Z ₀ (or an impedance-free generator which is in series with an impedance 2 Z ₀ ) is connected to the waveguide branch ι is. Since the waveguide branches 2 and 4 _{contain the same loads on Z 0} , no power will reach the waveguide branch 3 from the generator. Indeed, if this portion of the waveguide between points 2 and 4 is removed, operation will be essentially unaffected. This changed arrangement is shown in FIG. It is now clear that each arc of the ring is terminated with a characteristic impedance and that the generator of the impedance 2 Z ₀ is connected in series to two loads Z _0. The impedances are therefore matched to ensure maximum energy supply from the generator.

With the impedances corresponding to the arrangement according to FIG. 11, the behavior of the device is to be considered when the branch conductors at points 1, 3 and 4 are terminated with their wave resistances and when a generator of impedance Z _{0 is arranged} on the waveguide branch at point 2 is. Since the load impedances on the branches at points 1 and 3 are the same, the arc 3-4-5-6-1 contains only one standing wave and can therefore be removed from the ring. The corresponding arrangement is shown in FIG. It must now be taken into account, however, that the impedance, which through a one-quarter wavelength transmission line, whose characteristic impedance is Z ₀ , and which

Z ²

is terminated with the impedance Z, equal to —— ■

is. In this regard, the waveguides behave in exactly the same way. Each of the 2 Z ₀ loads according to FIG. 14 therefore appears as \ Z ₀ at the junction with the waveguide _{branch 2. Since the generator impedance Z 0 works} on two JZ ₀ loads in series, the impedances are matched and it exists maximum efficiency.

Other versions of double expansion joints

A series impedance Z on a transmission line can be replaced by a shunt impedance ™! -

be replaced, which is supplemented on both sides by a quarter-wave line. With the help of such a substitute, a duplex compensator with four series load impedances can be converted into other forms, the load impedances of which are arranged in various ways as series or shunt elements. However, when transitioning from series to shunt loads, it may be necessary to remove half a wavelength from any part of the ring in order to obtain the compensation characteristic. This substitute measure provides five different compensators in addition to the compensator of FIG. 11 discussed above. These additional compensators are shown schematically in FIGS. 15 to 18, each having a waveguide in the form of a closed loop or ring, with an average circumference of one and a half Wavelengths; In addition, four conductors are provided, which can be designed as hollow waveguides or coaxial cables, as shown schematically, and branch off from the ring at points that are distant from one another. The branch lines are connected to the ring either by means of a series connection or a parallel connection. Electrical distances between the branch lines along the ring in wavelengths which are determined by the type of connection, and also the characteristic impedances of the branch lines with respect to the characteristic impedance Z _{0 of} the waveguide ring, which are required for generating an impedance matching between the lines, are shown in the figures specified.

The operation of the shown in Figs. 15-18 Facilities can be analyzed using the same method that is related was used with the alternative device of Figure 11; it can be shown by the fact that every facility works in a way that corresponds to the desired behavior. The five in Figs. 11 and 15 to 18 devices shown can be used as be viewed fundamentally. Each of them can be made into an almost infinite number be made of different sized double expansion joints, with the following Rules to be followed.

1. Any integer number of wavelengths can be added to each arc;

2. Any whole number of half wavelengths can be added to any two arcs will.

These rules should apply in most cases find where it is desired to change the angles under the various connections, to simplify the construction or to

To obtain capacitors, which adapt more easily to the available space. It should, however it must be assumed that in general the lengthening of the arches is the practical range of the Operating frequencies reduced.

Since the dimensions of all double compensators are related to wavelengths, it is clear that every special design a more or less pronounced selectivity with regard to frequency

ίο has. At frequencies other than those The physical dimensions no longer remain on which the device is based just right, either in terms of creating the best impedance match or with reference to the best compensation. An explanation of the available bandwidths results from a consideration of special cases.

The basic circuits of FIGS. Ii and 17 achieve the compensation in that between opposite connections of the branch lines two paths can be created along the ring or loop, which geometrically around one half a wavelength differ from each other. If a frequency other than the design frequency is used, so these two paths no longer differ by exactly half a wavelength; a perfect one Compensation is not obtained and the miscompensation will increase as the difference between the operating frequency and the design frequency increases. A special model of any one therefore, these basic circuits cannot be designed to be perfect over a frequency band Compensation supplies. Examination of similar low frequency circuits reveals that in the case of small percentage deviations from the design frequency, the amount of incorrect compensation approximated by the equation

(fo) ²

is given, where P _{bp is} equal to the power developed in the load at the balance point (which should be the case), P _{1 means} the power developed in those loads which are to receive the power, / _{corresponds to the operating frequency, / 0} corresponds to the design frequency and C represents a constant on the order of one.

For frequencies commonly used in waveguides, the above expression shows that the models of Figures 11 and 17 can be used in many broadband applications. For example, if there is perfect compensation at / ₀ = 3000 megahertz, Pf, _{p will be} about 60 decibels below P ₁ at the edges of a 6 megahertz band and about 40 decibels below P ₁ at the edges of a 60 megahertz band.

In the basic circuits according to FIGS. 15 and 18 are the branch lines that lead into the ring at geometrically opposite points connected to the ring by means of a series connection and by means of a parallel connection. A burden that at one of these geometrically opposite series or parallel connections is arranged with Adjusted with respect to a transmitter attached to the other port. The right phase alignment or the correct compensation does not rely on the use of two different ways electrical length, but results from the inequality of the two types of connection. In addition, the compensation depends on the equality of the two loads that the Record performance. The degree of compensation that can be achieved at these two points is therefore independent on the frequency and dependent only on the extent to which the two loads can be made equal. The degree of compensation at the other two branch points depends on the frequency, in accordance with the approximate given above math expression. The degree of compensation at any point is common the only important criterion, and in many applications the arrangements according to FIG. 15 and 18 actually have compensation over a wide range of frequencies.

The double compensator shown in Fig. 16 compensates for any point due to the inequality of oppositely arranged connections. Even so, the compensation is frequency dependent. It does not matter where a transmitter is connected or where one Series loading and a parallel loading can be operated. Located at the design frequency virtual pistons in the ring, one at the port of the series load and one a quarter wavelength behind the connection of the parallel load. When the frequency changes these pistons are displaced unequal distances, and accordingly the loads appear no longer identical on the transmitter. As a result, the amplitudes aren't exactly the same, though two sets of waves with a phase shift of 180 ° arrive at the compensation point. It exists however, a partial compensation effect, which this double compensator a degree of compensation which is more constant with respect to frequency than in the cases of Figures 11 and 17.

With regard to the frequency dependency of the impedance matching, for all basic circuits the perfection of impedance matching will deteriorate when the operating frequency deviates from the design frequency. The loss resulting from the impedance mismatch grows but will usually be negligible. If, for example, a perfect adaptation for the model of FIG. 11 at a frequency of 3000 megahertz, then becomes the loss of 0.01 decibels caused by the mismatch at the edges of a 60 megahertz band do not exceed.

Two rules were given above which can be used to determine the size of any-

a! Compensator by adding wavelengths or half-wavelengths. One Another rule should be mentioned, namely that half a wavelength of any two is arbitrary Arches can be peeled off. However, this rule is only applicable to a limited extent, as it is only applicable to the Basic circuits according to FIGS. 15, 16, 17 and 18 are used can be. Concentrated circuits which apply the latter rule derived from the circuits of FIGS. 15, 16 and 18 are shown schematically in Figs. 19, 20 and 21; the circuits of FIG. 22 to 26 are all derived from Fig. 17, namely the first four, in that in different ways and way a pair of half-wavelengths have been removed, and the last by subtracting two Pairs of half-wavelengths. In all concentrated circuits there is a need superimposed connections. In the case of the superimposed connections shown schematically, the branch diagrams 7 to 10 made use of. It should be noted that the arrangements according to Figures 19-26 illustrate only one specific construction for each concentrated double compensator. Other measures to achieve the desired superimposed connections can be used naturally find application to realize other special arrangements which for certain Purposes may be superior to the explanations chosen for the purpose of explanation have been. It has already been indicated that some of the superimposed connections to a waveguide in practice there may be considerable deviations from simply superimposed connections. It is then possible that some of the concentrated embodiments of double compensators require extensive reactance tuning (not shown) to be operational to be.

As with the arrangements according to Figs. 15 and 18, so should the opposing branch lines with series and parallel connections too in the arrangements according to FIGS. 19 and 21 develop a compensation which depends only on the extent depends up to which two loads can be made identical. The frequency dependence the compensation in the remaining lumped models will be less serious than in their prototypes, but essentially the same Have character. It is also expected that all lumped circuits will have better impedance matching point over a given frequency band. For these reasons, the arrangements 19 to 26 to be particularly suitable in waveguide systems where compactness and a wide frequency response are essential. For other applications, of course, an extension of the Arches are preferable to facilitate mechanical construction.

Among the studied and tested models of double expansion joints were those 11, 19, 27 and 28. Fig. 29 shows the real one Construction used for the compensator sohematically shown in FIG. 27 and 28 differ from FIGS. 11 and 16 only in the addition of a whole Number of wavelengths to specific arcs, according to the first rule given above. It it was found that these models work exactly as assumed on the basis of simple assumptions where it was not necessary to apply a substantial correction factor to the first sample to remove the distortion of the electromagnetic fields caused by would have expected from ladder curvature and sharp bends.

It was found that in each of the experimental models the energy developed at a compensation point 45 decibels or maybe something was more below the energy sent to the ring. Such compensation is over a relatively wide frequency range achievable; however, the impedance mismatch changes over this area. The reason for this is that with the exception of a relatively narrow frequency band about the "design frequency," i. H. the optimal frequency, a certain reactance determination is necessary to obtain compensation through suitable phasing. This reactance determination however, the impedance of the double compensator changes when viewed from any series or parallel connection, and so the arrangement no longer match the impedance of the line to which it is applied.

Under reactance determination is either the introduction of suitable arrangements, which are adjustable Represent reactances in parallel with the resistance loads, in an or to understand several branches of the compensator or the careful detuning of the load itself so as to be a reactive component in the load impedance. It is of the greatest Importance that the reactance tuning has been determined experimentally feasible and that Due to the impedance mismatch, the operational attenuation for a wide frequency range is not needs to go beyond a few percent of the energy.

It should be noted that those shown in FIGS. 11, 27, 28 and 29 The experimental models illustrated form typical patterns of the various theoretical circuits. In terms of the excellent performance of each of these practical designs, the Problem of refinement of the execution for the purpose of improving the degree of compensation and for the purpose of Reduction of the operational damping no difficulties. In fact, this would only be considered a design problem to be considered, which only with reference to specific sizes and shapes of the waveguide and special wavelength ranges can best be solved.

Double expansion joints using other types of transmission lines

The general principles of which the mode of action of the double-

409 765/5

compensators depends; are not the waveguides alone peculiar; the basic circuits shown in Figs. π and 15-18 Rather, they can apply to the application of any type of ordinary "return route" transmission line, for example a coaxial cable or a shielded twin-core cable (parallel wire lines) or to the use of any suitable combination thereof, including the ίο Waveguides ■ to be set up. In the specification a basic double compensator circuit for each type of transmission conductor and for each Type combination must be ensured that the correct line and load impedances are provided and that according to the respective requirement, a series or parallel connection is carried out.

It should be noted that certain combinations of different types of transmission lines with each other and with waveguides are easier to handle than other combinations. For example, can It is easier to use coaxial cable in connection with waveguides than a shielded double (parallel wire) cable; this applies to the mechanical design of series and parallel connections as well as the difficulty of electrical symmetry and impedance matching realize. Furthermore, it should not be assumed that in general there is a given ■ 3 ° conductor type with advantage for the entire double expansion joint can be used. For example, a series connection of coaxial cable can be used much more convenient to run from a shielded pair of cables than from any other coaxial cable. However, in most cases it is likely to be desirable to have only one type of pipe for expansion of the "ring" part of the compensator.

Combinations of waveguides and transmission lines, which is the most practical construction are summarized below:

ring
consists Simplest
Series connection
with Simplest parallel
connection with Waveguide
Coaxial cable
shielded
Pair Waveguide
shielded
Pair
shielded
Pair Coaxial cable
shielded
Pair
shielded
Pair

It should be noted that in the double compensators according to the invention, which correspond to a waveguide ring Figs. 11 and 15-16 have the branch lines, which are connected to the ring by means of a parallel connection, as coaxial lines as the use of waveguides for parallel connections is likely it is impractical unless the ring comprises many half wavelengths, i.e. one large one Has a radius of curvature, or if the ring is not so deformed that it has one or more has straight parts. . "..

The basic circuitry for a ring of coaxial cable is shown in Figures 30-34 and for a ring of shielded pair wire is shown in Figures 35-39. Only the branch line connections as listed in the table have been shown. As shown, the impedance values and dimensions for the coaxial cable twin compensators of FIGS. 30 through 34 and for the shielded wireline compensators of FIGS. 35 through 39 are the same as for the corresponding compensators having a waveguide ring of FIGS 18 use. The lumped circuits derived from the basic circuits of FIGS. 15 to 18 and shown in FIGS. 19 to 26 are of course applicable to the double compensators of FIGS. 30 to 39.

Double expansion joints with lumped constants

A section of transmission line of any length can be used for any particular operating frequency by a symmetrical T or π network with concentrated circular constants and the impedance values necessary for the display are well known. To convert the transmission line double compensators as described above into double compensators to convert which circles with lumped constants use, d. H. Wash, Capacitors and resistors, it is only necessary to concentrate the matching T and π sections with Constants instead of half-wave, quarter-wave, and three-quarter-wave segments to be arranged in the ring or loop between the branch lines of the above Compensators were used. To represent the power dissipation of the real Lines is usually a series resistor in each branch of every T and π network added. However, coils and capacitors always have power loss, so that for the present Purpose the assumption is particularly simple that the losses in the coils and capacitors are set to accurately represent a given attenuation line.

The representation of quarter and three-quarter wavelength segments of circles with lumped constants is straightforward; there is, to be sure, an ambiguity regarding the Sign of the reactances; but in any case this can be easily done by a current analysis to solve. The case of a half-wavelength section is not so simple. The analytical solution based on Impedance based only results in one pair of wires. The analysis, which is based on currents, shows, however, that these wires must cross. Nonetheless, even crossed wires cannot can be freely used as a substitute for a half-wavelength section. Generally they are crossed Wires entirely inadequate for the behavior of a half-wavelength segment with reference

to represent the frequency. In the present case there is also the double compensator with lumped constant which is derived from the arrangement of Fig. 17 by three pairs of crossed wires applied, no compensation at all. It is therefore necessary Be careful when displaying half-wavelength sections. It has been found that two T or π parts with concentrating th constants which are in series and each of which is a quarter or three-quarter wavelength segment is equivalent to give a fair representation in cases where crossed wires fail. It can also be shown that the network representation of quarter and three-quarter wavelength segments for small deviations from the given frequency the exact functional change unfolds, which occurs from the expression of the transmission line only by numerical constants of the order of magnitude Plins is different.

Using symmetrical electrical T or π networks or their combinations of correct equivalent wavelength in place of the corresponding waveguide lengths in the Waveguide double compensators according to FIGS. 11 and 15 to 18 can have a large number of double expansion joints can be derived with lumped constants. This number can still be passed

jo application of the rule that increase the representation of half a wavelength with lumped constants can be added to any of any two arcs. To illustrate this, the Circuit diagrams of six such double compensator

! 5 ren with lumped constants derived from Fig. 11 and 15 to 18 are shown in Figs. 40 to 44.

Applications of double expansion joints

to As already said, the double expansion joints are according to the invention primarily of value for double messaging systems because they allow a simultaneous transmission of messages in two directions at the same frequency. To between

■ 5 see two stations A and B setting up communications, either (a.) Transmission lines or (b) a radio path can be used to connect the stations. In both cases the double expansion joint can be used with

ο be used to great advantage. In Fig. 45, there is shown a schematic circuit diagram in the case of a transmission line. Here the use of a double compensator of any of the types described above (shown schematically in Fig. 45 and in the following figures) with the receiver arranged in an arm compensated with respect to the transmitter enables simultaneous transmission and reception at each station with only one line between A and B. In each transmission there is a loss of 3 decibels due to the power dissipation in the reactive load, and upon reception there is a loss of 3 decibels in the transmitter. The total loss of 6 decibels will in some cases be of no consequence compared to the economic benefit of only requiring half the length of the transmission line; in other cases the single line itself can be improved to compensate for the 6 decibel loss with economics of space and cost or both.

46 is a schematic diagram for the case of radio communication over the air. The presence of a double compensator in both stations A and B enables the arrangement of a single antenna or radiation system, which can be used for reception and transmission at the same time. As in the case of Figure 45, there is a total loss of 6 decibels when messaging in both directions. In directional antenna systems, however, the space available for the radiator is often limited. The use of a double compensator allows the entire space to be used for transmission and reception, with a gain of 3 decibels for each operation. The power levels for the arrangement according to FIG. 46 with a "double-sized" radiation system are therefore the same as for a system with two separate antennas of one size for reception and transmission.

In its simplest form, an object location system consists of a transmitter, which operates a directional antenna, with short or ultra short waves being transmitted, and from one Receiver that is connected to a nearby directional antenna and through which that of an object reflected waves are received. Fig. 47 shows schematically how a double compensator according to the invention can be used for object location with a single emitter. Of the A total loss of 6 decibels can be achieved by doubling the "size" of the radiator (in the case of a Hornes by doubling the opening size) can be precisely balanced. Such a radiator double size, however, has a considerably greater "angular directivity" than a simple radiator Size; and such increased angular resolution is often very desirable. The order According to FIG. 47, object location can then be used as special for many purposes prove efficient.

It might seem like the total loss of 6 decibels in duplex operation is unavoidable as long as passive transmission elements are used. But it can be said in advance that for some cases of duplex transmission, the circuit elements are set to sharp pulses with the transmitter in a dormant state between pulses. In this state of affairs the use of active circuit elements is recommended, which creates other arrangements the loss inherent in current double expansion joints of 6 decibels off. It should be noted, however, that the double compensator with active elements in the transmitter and appendix arms would show a total loss of only about 3 decibels.

In connection with problems of object location, the question arises what kind of polarization should be used for sending and receiving. Linearly polarized waves are the easiest apply, but there may be cases in which the polarization angle is chosen very unfavorably is with respect to some objects in certain positions, thereby reducing the reflected energy is too small to be picked up. It is ίο clear, however, that the use of circularly polarized Waves solves any such difficulty.

In the event that circular polarization is used, it is important to know whether a waveguide double compensator is readily available. It should be remembered that the branch in the electrical plane is a series connection and in the magnetic plane a parallel connection. If, therefore, the circuit according to FIG. Ii is carried out with a circular waveguide for the ring and for all branch arms, a compensator is created for linearly polarized waves with an electric field strength in the plane of the paper and for linear polarization at right angles to the plane of the paper. Since the circuit behaves correctly for both polarizations, it represents a double compensator for circular or elliptical polarization; however, care must be taken to ensure that the correct impedances are provided separately for the two perpendicular components. This should not be difficult, especially when the two components are received separately to form a polarization multiple receiving system. Even if all impedances remained the same as Z ₀ , the loss due to the impedance mismatch would not exceed 1 decibel. It is known from ordinary telephone practice that through the use of devices which operate in a manner similar to a voice coil, the construction of a two-way amplifier is a straightforward matter. Two double compensators according to the invention can therefore be used in conjunction with two amplifiers to form a double amplifier. Fig. 48 schematically shows such a waveguide amplifier using waveguide amplifiers with speed variation. It is characteristic of these amplifiers that the feedback between the final output gap and the previous gap is very small. It follows that the feedback from this amplifier is very little.

49 shows a way in which any three double compensators according to the invention, as described above and shown schematically in the figure, can be connected directly in order to form a "multiple double compensator". Such a would be of great use in the case where one station 0 to four stations A, B, ₁ C, D should transmit simultaneously and station 0 should receive from station A or B and C or D at the same time. Even more double expansion joints can be linked directly in a similar manner to daisy chain a greater number of stations; z. B. seven compensators can be used for a central station and eight sub-stations. In such an application there is no loss in transmission and no more than 3 decibels loss in reception; the use of radiators or pipes of double size therefore leads to a gain of 3 decibels for such arrangements.

In general it can be stated that the double compensators according to the invention are advantageous in most cases can be used where signals of the same frequency, which in are transmitted in the opposite direction on the same line or by air, should be separated. The total loss associated with such double messaging methods of 6 decibels must be accepted and endured, although this loss, as above was mentioned, often reduced or disappeared due to some changes in the plant can be brought.

It should be noted that in the above description of the double compensators according to the invention one the same energy distribution has been assumed for the two loads operated. If the Arrangements constructed in accordance with the above explanations is a uniform one Energy sharing necessary to ensure compensation. With appropriate modifications however, compensation can also be obtained without an even distribution of energy. if Cases arise whose requirements are best achieved with a compensator with unequal energy distribution can be satisfied, the double compensators already described must be determined by reactance be changed within the ring or by using a ring, composed of two or more suitably coupled transmission line arcs which differ from one another in their type as much as possible and have different wave impedances exhibit. 10:

In the general case, one load can then absorb p percent of the energy emitted, while the other load receives (100- p) percent. In the case of passive circuit elements, reciprocity requires that when the transmitter (signal generator) is connected in place of the first load, p percent of its emitted energy must be delivered to a suitable load located at the original transmission point. Since the product p (loo-p) is a maximum for ^ = 50 percent, it can be seen that the case of uniform energy distribution has the highest double efficiency. In view of this efficiency consideration and in view of the fact that an unequal energy distribution would result in considerable additional complications in the construction and operation of the arrangement, the case of equal energy distribution is normally preferred and has accordingly been considered for the above explanations alone.

For those skilled in the art, there will be many changes beyond the changes explained and described give further options for changing the training and application of the instructions, but without thereby deviating from the essence of the invention and the scope of the invention is exceeded.

PATENT CLAIMS:

i. Coupling arrangement for the coupling of four transmission circuits A, B, C and D of a wave transmission network as branches to a closed transmission loop, the electrical length of which is an odd multiple of a half-wave length at the operating frequency, such that the wave transmission between each circuit of a first, the circles A and C and each circle of a second pair comprising circles B and D can take place, but that the transmission of waves between circles A and C or B and D of the same pair is largely prevented, characterized in that using either one Parallel or a series connection for each branch at least one of the branches is connected to the closed loop by means of a series connection, so that the electrical distance existing along the loop between two branch circuits connected next to one another is approximately zero or a whole multiple (o, i, 2, 3, 4) the fourth elwavelength at operating frequency and the distance along the loop between the two circles of each pair approximated to zero or an even multiple (0, 2, 4, 6. . .) The quarter wavelength is dimensioned at the operating frequency if the connections to the loop are of the same type, ie both series or parallel connections, that the distance between the branches of a pair along the loop when using a series connection for one branch and a parallel connection for the other branch is approximately an odd multiple of a quarter wavelength, and that at most three of the four branches are connected at any given point on the loop.

2. Coupling arrangement according to claim 1, characterized in that the transmission loop has the same characteristic impedance over its entire length and that each circuit of the one pair has an impedance equal to the characteristic impedance at its loop connection point of the loop circuit, while the impedance of the two circuits of the other pair is dimensioned at their loop connection points so that the sum of their impedances, through seen the loop circle over each of the first mentioned circles, also equal to the wave resistance of the loop circle is.

3. Coupling arrangement according to claim 1 or 2, characterized in that the Loop circle consists of a coaxial line.

4. Coupling arrangement according to claim 3, characterized in that at least one of the branch circuits connected to the loop is designed as a coaxial line, the inner and outer conductor to form a parallel connection with the respective ones Ladders connected to the loop.

5. Coupling arrangement according to claim 3 or 4, characterized in that at least one of the branch circuits connected to the loop from a shielded two-wire line which is connected in series with the inner conductor of the loop.

6. Coupling arrangement according to claim 1 or 2, characterized in that the Loop circuit consists of a shielded two-wire line.

7 · coupling arrangement according to claim 6, characterized in that each of the Circuits connected to the loop circuit is designed as a shielded two-wire line.

8. Coupling arrangement according to claim 1 or 2, characterized in that the Loop circle from a ring-shaped waveguide for polarized electromagnetic waves exists whose electrical vector runs parallel to the loop plane.

9. Coupling arrangement according to claim 8, characterized in that at least one of the circles connected to the loop consists of a waveguide which is essentially extends radially to the annular waveguide loop and its electrical plane for the purpose Forming a series connection is parallel to the electrical plane of the loop.

10. Coupling arrangement according to claim 8 or 9, characterized in that at least one of the circuits connected to the loop consists of a coaxial line, the inner Conductor leads into the interior of the annular waveguide for the purpose of forming a parallel connection.

11. Coupling arrangement according to claim 1 or 2, characterized in that the loop circle consists of two or more in series lying networks with lumped circular constants, each of which is essentially a transmission line of a quarter wavelength at the predetermined operating frequency.

Documents considered: German Patent No. 540 615; U.S. Patent Nos. 1,944,283, 2,134,278, 147809, 2244756;

Proc. of the I.R.E., 1941, pp. 115 to 120.

For this purpose 4 sheets of drawings

© 409 765/5 12.

Claims (1)

Supplementary sheet to patent specification No. 977 019 Class 21a * Gr. 74 ISSUED MARCH 25, 1965 Patent 977 019 is by a final decision of the German Patent Office
from December 31, 1964. a) The previous description page ι to page 2, Line 28 has been changed as follows: The invention relates to a coupling arrangement for coupling four transmission circuits A, B, C, D of a wave transmission network. It is known to couple four transmission circuits as branches to a closed transmission loop, the fixed electrical length of which is an odd multiple of a half-wavelength at the operating frequency, in such a way that the wave transmission is continuous between each individual circuit A or C of a first, the circuits A and C and both circles B and D of a second pair, including circles B and D , and equally between both circles A and C of the first pair and a single circle B or D of the second pair, but that the wave transmission between the circles can take place A and C or B and D of the same pair is largely always prevented. In the known coupling arrangement, all connections are designed as parallel connections. The distance between two branches of different pairs is a quarter of the wavelength at the operating frequency or an odd multiple of the quarter wavelength, during the Distance along the loop between the two circles of each pair is an even multiple of Quarter wavelength is. Resulting from the use of parallel connections for all branch circuits certain restrictions in use; In particular, the known arrangement can be used as Use double waveguide compensator only in such a way that a conversion occurs in one side of the loop or interchanging the conductors is carried out in order to delete the two sides of the loop at the opposite connection to realize propagating wave energy. The invention aims to perfect a coupling arrangement of the type indicated, and in particular in the implementation of double compensation, the dependence on conversion or exchange measures avoid. The special feature of the invention is that using either a parallel or a series connection for each branch at least one of the branches is connected to the closed loop by means of a series connection, so that the fixed electrical distance between two branch circuits {AB, BC, CD, DA) of different pairs for their mutual coupling in the presence of different types of connections in the circles under consideration is measured with approximately zero or an even multiple of the quarter wavelength, on the other hand, when similar connections are present, approximately with an odd multiple of the quarter wavelength, the latter Dimensioning rule for parallel connections is known that the existing along the loop distance between two branch circles of the same pair (AC or BD) for their mutual decoupling in the presence of different types of connections of the circles under consideration with approximately one Odd-numbered multiples of the quarter wavelength is measured, however, if there are similar connections with approximately zero or an even multiple of the quarter-wavelength known for parallel connections, and that at most three of the four branches are connected to the same point of the loop. b) Claim 1 has been amended as follows: Coupling arrangement for the coupling of four transmission circuits A, B, C and D of a wave transmission network as branches to a closed transmission loop, the fixed electrical length of which is an odd multiple of a half-wave length at the operating frequency, such that the wave transmission is continuous between each individual circuit A or C. a first pair comprising circles A and C and both circles B and D of a second pair comprising circles B and D and equally between both circles A and C of the first pair and a single circle B or D of the second pair can ^ that [but the wave transmission between the circles A and C or B and D of the same pair is largely always prevented, characterized in that using either a parallel or a series connection for each branch at least one of the branches by means of a series connection to the closed loop is connected, that the existing fixed electrical distance along the loop between two branch circuits (AB, BC, CD, DA) of different pairs for their mutual coupling in the presence of different connections b ^J ei the circles under consideration with approximately zero or an even multiple the quarter wavelength is measured, on the other hand, if similar connections are present, approximately with an odd multiple of the quarter wavelength, whereby the latter rule for parallel connections is known that the distance between two branches of the same pair (AC or 5D) ^{along 1 of the loop changes} lateral decoupling when there are different types of connections of the circles under consideration with approximately an odd multiple of the quarter wavelength, on the other hand when similar connections are present with approximately zero or an even multiple of the quarter wavelength known per se for pair connections, and that at most three of the four branches on connected to the same point on the loop.