DE912235C - Electrical waveguide system for creating a phase difference between two output waves - Google Patents

Description

The invention relates to an electrical waveguide system for creating a phase difference between two output shafts and to a device with such a waveguide system.

The invention is based on an arrangement consisting of two waveguides of rectangular cross-section, of which the first has a length k, a width a and a height b, and the second has a length / deviating from k , a width c and a height d . In-phase waves are fed to these waveguides on the input side.

The waveguide system according to the invention has the feature that the value of α and c both at the entrance and at the exit of the waveguide in question is equal to a ₀ and α or c for at least one of the two waveguides changes continuously over its length and has an extreme value between the input and the output, that (a - a ₀ ) or (c - a ₀ ) for this extreme value is small compared to a ₀ and that (O ₁ -O ₂ ) <3 « ₀ (k - /), so in what

O ₁ = f [a - a ₀ ) ds and O ₂ = Γ (c - a ₀ ) ds.

t - 0 S = O

The invention is illustrated below with reference to the drawing explained in more detail.

Fig. Ι shows a known waveguide system;

FIG. 2 is a graph comparing the operations of the waveguide systems of FIG Fig. Ι and after Fig. 3;

3 illustrates a waveguide system according to one embodiment of the invention.

In waveguide technology, a phase difference is sometimes to be created between two waves. In a push-pull crystal modulator z. B. this phase difference must be 180 ° and 90 ⁰ with a push-pull transmit / receive switch. It is known that any phase difference can be produced by the waveguide system of FIG.

This system consists of the waveguides 1 and 2. The waveguide 1 has a length k and a rectangular cross section with the width a ₀ and the height b. The waveguide 2 has a length I and a rectangular cross section with the width a ₀ and the height d.

Both waveguides have a constant cross section over their entire length. Such rectangular waveguides are generally used for waves with a wavelength A, the frequency of which is somewhat above the lowest critical frequency. These are waves of the type If ₁₀ , in which a so-called TE vibration type occurs. In particular, A is chosen in such a way that waves with other types of oscillation do not yet occur. This is the case when a _Q <A <2 a _a . A wave with a wavelength A ₃ occurs in the waveguide, which, as is well known, is not equal to A, but rather greater

\ ^2a o)

A is ₀ <A <2 a _0, then always A _{3> ^2/3} a ₀ J / 3 system. If waves with the same phase are fed to the input sides 3 and 4 of the waveguides ι and 2, waves can be found on the output sides 5 and 6 of these waveguides which have a certain phase difference φ . As can easily be seen, this is the phase shift

h-l

360 ⁰ .

(2)

With the correct choice of k and I , any desired phase shift can be achieved. The phase shift should often not be dependent on the frequency or on the wavelength. With the push-pull crystal modulator, for example, a constant phase shift of i8o ^{c is} required in a certain frequency range. Apparently the system according to FIG. 1 does not meet this requirement. From (1) and (2) it follows that if A changes, A ₃ and thus φ also change. This behavior is shown by line A in FIG. As an example, a system has been chosen for which a _Q is 2.29 cm and k - I is 1.19 cm. For a wavelength A = 3.32 cm (A ₃ = 4.76 cm), a phase rotation φ of 90 ⁰ occurs. For larger values of A and A _s , φ is smaller and vice versa, as line A also shows.

In the waveguide system according to the invention, which is shown in one embodiment in FIG. 3, this disadvantage is avoided. Corresponding parts are denoted by the same reference numerals. Both waveguides here have a width a ₀ near the ends. The width α of the waveguide 1 has a maximum between the ends as a function of the length s, while the width c of the waveguide 2 has a minimum as a function of s.

It can now be seen that the phase shift ψ between the waves that can be removed at the output sides 5 and 6, when in-phase waves are fed to the input sides 3 and 4, is practically no longer dependent on the wavelength in a certain wavelength region. This wavelength range will be in the above-specified range where

«O <A <2 a ₀ ,

and so be useful in this system if

(O ₁ -O ₂ X ₃ a ₀ (k-1),

(3)

worm

O ₁ = J (a - a _Q ) ds and O ₂ = f (c - a ₀ ) d s.

«= O

It turns out that O ₁ and O _{2 have} a simple geometric meaning. They represent the enlargement of the surface of the waveguide, which is due to the inconsistent width. For example, for waveguide 1, O _{1 is} equal to the sum of hatched surfaces 7 and 8. For waveguide 2, the sum of hatched surfaces 9 and 10 is equal to —O ₂ . The minus sign is due to the fact that the surface has become smaller here than with a constant width a ₀ .

It turns out that the center of the wavelength range, in which φ practically does not depend on the wavelength, lies at

TI TV

- 0,

The phase shift that occurs is the same

! / (O _1. -O ₂ ) (A -i)

φ = 36o °

(5)

If a phase shift of η ■ go ^{c is} desired, where η means an integer, which will often occur, then the expression

O ₂ ) (k-1)

practically be an integer.

From the formulas (4) and (5) the condition for the choice of k - I results

k - l

720 °

(6)

In a waveguide system according to the invention, half of the phase rotation can thus be thought of as being caused by the longitudinal difference (k · −I) and the other half as being caused by the effect of the changes in width.

In FIG. 2, again for the case a ₀ = 2.29 cm, the course of φ is shown as a function of A and X ₉ for the waveguide system according to the embodiment of the invention under consideration. The center of the area in which φ remains practically unchanged is placed at λ _β - 4.76 cm. This through the

Line B is essentially horizontal, which means that φ here remains practically independent of λ and Xg equal to 90 ^0.

One of the surface parts O ₁ and O ₂ can assume the value zero without hesitation, that is to say that one of the two waveguides 1 and 2 can have a constant width a ₀ .

It should be noted that the dimensions b and d only need to meet the requirements normally imposed on waveguides.

In Fig. 3 the changes in width are exaggerated for the sake of greater clarity.

It can be seen that where protruding from width the

The talk was that the effective width should be understood.

All known measures for influencing the width are within the scope of the invention to achieve the Changes in width applicable.

Claims (3)

Patent claims: ao i. Electrical waveguide system for generating a phase difference between two output waves consisting of two waveguides of rectangular cross-section, the first of which has a length k, a width α and a height b and the second has a length I, a width c and a height d different from k , the In-phase waves are fed to the input sides of the waveguide, characterized in that the value of α and that of c both at the input and at the output of the waveguide in question is equal to a ₀ and α and c for at least one of the waveguides extend over its length changes and between the input and the output has an extreme value that (a - a ₀ ) or (c - a ₀ ) for this extreme value is small compared to a ₀ and that (O ₁ - O ₂ ) <3 « ₀ [k - l), where O ₁ = f (a - a ₀ ) ds and O ₂ = J (c - a _o ) ds. S = O »= 0 2. Waveguide system according to claim 1, characterized in that to achieve a phase difference of 90 ⁰ or a multiple thereof, the dimensioning is chosen such that the value of h - Qj) {k - i) al
is an integer. 3. Electrical device with a waveguide system according to claim 1 or 2 and a generator of electrical waves connected to both input sides of the waveguide, characterized in that the generator generates waves with a wavelength λ which at least essentially meets the condition: -i) -Oo Fulfills. 1 sheet of drawings © 9504 5.54