JPS6229921B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a low-pass filter circuit used in frequency bands such as EHF, SHF, UHF and VHF. Conventionally, in low-pass filter circuits in which distributed constant lines with different characteristic impedances are alternately connected in cascade, design values based on distributed constant theory are required due to complex electromagnetic effects caused by sudden changes in characteristic impedance at these connections. First, we had to make a prototype and then go through trial and error to obtain the desired filtering characteristics, which was complicated. The present invention provides a tapered shape having a gradual change in the characteristic impedance value of one distributed constant line to the characteristic impedance value of the other adjacent distributed constant line at the connection portion of the distributed constant lines of the low-pass filter circuit. Alternatively, the filter is characterized in that minute step-shaped lines are provided and connected in cascade to each other, and the purpose is to easily obtain desired filtering characteristics without trial and error based on theoretical design values. This will be explained in detail below with reference to the drawings. FIG. 1 shows one section of a low-pass filter circuit composed of distributed constant lines. The transfer matrix that expresses the transmission characteristics of this section is the speed of light, c, frequency, and the free space line length of the distributed constant line (line length converted when relative permittivity εr=1).
_i , the terminal impedance of the low-pass filter circuit is R, and the characteristic impedance of the distributed constant line is Z _0i , However, it is expressed as θ _i =2πl _i /c. Further, the transmission characteristic of a low-pass filter circuit in which N sections each having line constants Z _0i and l _i are connected in cascade is represented by the product of all the transfer matrices of each section. The order of multiplication of the transfer matrix is performed in accordance with the order of each section of the distributed constant line cascaded from the input side to the output side. The total transfer of N sections obtained in this way is
Each component of the matrix is T ₁₁ ^l , T ₁₂ ^l , T ₂₁ ^l and
Let T ₂₂ ^l . Next, the transfer matrix is converted into a physically meaningful scattering matrix (S matrix). Let each component of the corresponding N-interval total scattering matrix be S ₁₁ , S ₁₂ , S ₂₁ , and
S ₂₂ , their relationship is It is indicated by. Therefore, the input loss from the input side to the output side of the low-pass filter circuit is 20log | S ₂₁ | = -
20log | T ₁₁ ^l is determined in decibels, and the reflection coefficient on the input side is 20log | S ₁₁ | = 20 log { | T ₂₁ ^l | / | T ₁₁ ^l
｜decibel. By calculating the input loss and reflection coefficient for each frequency, the frequency characteristics of the low-pass filter circuit can be determined. Here, we have given the characteristic impedance Z _0i and free space line length l _i of each section of the low-pass filter circuit, and have shown a procedure for calculating the frequency characteristics. However, in reality, the input loss for the frequency of the intended low-pass filter circuit,
Z ₀ of each distributed constant line section to realize reflection coefficient etc.
Since it is necessary to determine _i and l _i (i=1, 2, . . . N), the procedure is shown below. Find the square of the difference between the input loss obtained by giving the initial values of Z _0i , l _i (i=1, 2,...N) for each interval and the desired input loss for each frequency, and calculate the sum of them. Let be Q. By minimizing Q by using Z _0i and l _i (i=1, 2, . . . N) of each section as variable parameters, a desired low-pass filter circuit can be designed. The method of determining Z _0i and l _i (i=1, 2,...N) in this way is called an optimization method, and the parameter Z for each interval is determined by repeatedly performing calculations using an electronic computer using a mathematical method. _0i , l _i (i=1, 2, . . . N) can be obtained. An example of a conventional analytically designed low-pass filter circuit is shown in FIG. In the figure, Z _0i and l _i indicate that seven sections of distributed constant lines with different values are connected in cascade alternately, and each section is designated as B, D, F,
They are designated H, F', D' and B'. It also schematically shows that the characteristic impedances of adjacent sections are significantly different. As mentioned above, a circuit including a cascade connection of distributed constant lines with significantly different characteristic impedances involves electromagnetic effects that are outside the range considered in the distributed constant line model in Figure 1, so the required transmission characteristics cannot be achieved. Therefore, it was necessary to modify the design values through trial and error. In the present invention, in order to prevent the characteristic impedance of adjacent distributed constant lines from changing suddenly, a stepped or tapered line portion is provided that gradually changes in appearance.
Each distributed constant line constant Z _0i , l _i (i=1, 2,...
N) is optimally determined. As an example, FIG. 3a schematically shows a low-pass filter circuit with a terminating resistor of 50 ohms and an input loss of 0.3 dB or less in a frequency band of 1050 MHz or less. In the same figure, sections a, a′c, c′, e, e′, g and
g' section indicates a stepped track section, c, c', e,
Since e', g, and g' are each subdivided into four sections, the total number of sections N is 39. FIG. 3b is a schematic diagram of the case where the envelope of the stepped line section of FIG. 3a is averaged and tapered. Further, FIG. 4 is a diagram illustrating the distributed constant line configuration of FIG. 3a,
1, 2, 3, . . . 39 are line numbers assigned sequentially from the input side to the output side. Table 1 shows line constants Z _0i , l _i (i=1, 2,...
39). Further, Table 2 shows the frequency characteristics of the input loss and input reflection coefficient of the low-pass filter circuit of FIG. 4, that is, FIG. According to this 1050M
It can be seen that the input loss increases from Hz to 1450 MHz and has low-pass filtering characteristics. Further, similar characteristics can be obtained even if the shape is tapered. It should be noted that when a connection portion in which the characteristic impedance does not change significantly is provided as a component of a low-pass filter circuit, it can be easily inferred that the connection portion does not have to have a stepped or tapered shape.

【table】

【table】

【table】

[Table] As explained above, the present invention provides a low-pass filter circuit with a tapered or stepped shape that gently changes the characteristic impedance from one characteristic impedance to the other at the connection portion of each distributed constant line. Since it is configured by providing a line, it has the advantage that desired filtering characteristics can be easily obtained without trial and error.

[Brief explanation of the drawing]

Fig. 1 is a schematic diagram showing one section of a distributed constant line, Fig. 2 is a conventional low-pass filter circuit, and Fig. 3a is an embodiment of the present invention. b shows another embodiment of the present invention in which a tapered line is provided. FIG. 4 is an explanatory diagram of FIG. 3a. 1 to 39 in FIG. 4 indicate line numbers.

Claims (1)

[Claims] 1. In a low-pass filter circuit in which distributed constant lines with different characteristic impedances are alternately connected in cascade,
A step-like or tapered line having a gradual change from the characteristic impedance value of one distributed constant line to the characteristic impedance value of the other adjacent distributed constant line is provided at the connection part of these distributed constant lines. Features a low-pass filter circuit.