JPH05129803A - Switch - Google Patents

Abstract

PURPOSE:To improve isolation between input and output terminals by multilayering a semiconductor integrated circuit so as to prevent adverse influences such as inter-line coupling at a microwave semiconductor switching matrix. CONSTITUTION:A switching function formed on a semiconductor layer 7 is obtained by providing the three layers of a first layer 13, second layer 12 and third layer 7 composed of dielectrics or semiconductors, forming a one-face grounding conductor 14 between the adjacent first and second layers 13 and 12 in the three layers and independently constituting two microstrip lines while using the first and second layers 13 and 12. The gate of a field effect transistor 9 is connected through a bias hole 17a to an input line 2d, the drain is connected to an output line 4d and the source is connected through a bias hole 17b to a common ground conductor 14. The thickness of the ground conductor 14 is decided more than thickness so as to prevent surface currents generated on both sides of the ground conductor 14 from interferring each other.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to improvement of isolation between input and output terminals of a microwave switch matrix capable of outputting a microwave signal incident from a plurality of input terminals to a plurality of arbitrary output terminals. It is a thing.

[0002]

2. Description of the Related Art Conventional Example 1. FIG. 7 shows, for example, S. Powell,
■ Broad Band Monolithic Cross Point Switch Matrice
sIEEE Microwave and Millimeter-wave Monolithic C
It is a block diagram of the conventional microwave switch matrix shown by ircuits Symposium1990 pp.127-130. Generally, a switch matrix is composed of M input terminals and N output terminals, but this conventional example shows a case where the number of input terminals is 2 and the number of output terminals is also 2. In the figure, 1 is an input terminal, 2 is an input line, 3 is an output terminal, 4 is an output line, 5 is a switch, and 6 is a terminating resistor.

Next, the operation will be described. Input terminal 1
The microwaves incident on one or both of a and 1b are
It propagates through the input lines 2a and 2b. At this time, if the switch 5 provided near the intersection of the input line 2 and the output line 4 is closed, a part of the microwave is split to the output line 4. The demultiplexed microwave propagates through the output line and is output to the output terminals 3a and 3b.

FIG. 8 shows the operation state more specifically. In this case, since the switches 5a and 5b are both closed, the microwave incident on the input terminal 1a is not transmitted to the output line 4
The signals are demultiplexed into a and 4b and output to both output terminals 3a and 3b. On the other hand, since the switches 5c and 5d are open,
The microwaves incident on the input terminal 1b are output lines 4a, 4
It is not demultiplexed in b and is absorbed by the terminating resistor 6 and does not appear in the output terminal 3.

Next, a specific structure will be described.
FIG. 9 is a perspective view showing the structure of the switch 5 and the intersection, which are important for switching the input / output paths of the switch matrix. The input line 2c and the output line 4c are configured by the microstrip line using the semiconductor substrate 7. An air-bridge 8 is used in the middle of the output line 4c so that the input line 2c and the output line 4c do not come into contact with each other at the intersection. Further, a field effect transistor 9 is provided for connecting the input line 2c and the output line 4c. The gate and drain of the field effect transistor 9 whose source is grounded are connected to the input line 2c and the output line 4c, respectively. Here, the field effect transistor 9
A bias applying circuit is further required for the operation of, but its illustration is omitted here. Reference numeral 10 is a through hole for connecting the source of the field effect transistor 9 to the ground conductor 11 and grounding. Field effect transistor 9
If the applied bias of (1) is set to a condition similar to that of a normal amplifier, the input line 2c and the output line 4c are connected. At this time, due to the directivity of the field effect transistor 9, the output line 4
Reverse wave propagation from c to the input line 2c is blocked. On the other hand, when the applied bias to the gate is deepened to set the pinch-off voltage, the field effect transistor 9 does not operate,
The input line 2c and the output line 4c are cut off.

Conventional example 2. FIG. 10 shows another conventional microwave switch matrix mounting circuit, in which a unit circuit is formed for each crosspoint portion and a plurality of unit circuits are combined and interconnected to form a predetermined m × n matrix. It is designed to form a matrix. That is, in the cross point portion, a pair of first and second dielectric substrates 27 and 28 in which a micro circuit is formed so as to prevent coupling due to the intersection of lines are laminated with the ground conductor 29 interposed.
Of these, the first and second input terminals 10 and 11, the input coupler 23, the microwave switch 25, and the 50 Ω termination 26 are formed on the first dielectric substrate 27, and the first and second input terminals 10 and 11 are formed on the second dielectric substrate. To the first and second output terminals 20, 2
1 and an output coupler 24 are formed. The first and second dielectric substrates 27 and 28 are laminated so that the microwave switch 25 and the output coupler 24 are attached to each other with the coaxial feed through 30 electrically connected to each other. Next, in the mounting circuit configured as described above, 2 × 2 matrixes are formed by combining and connecting, for example, 2 × 2 pieces according to the scale of the matrix, and m ×
An n × n matrix is formed by combining and connecting n pieces.

[0007]

Since the semiconductor integrated circuit of Conventional Example 1 is constructed as described above, it is necessary to prevent the contact between the input line and the output line provided on the same plane, and therefore, at the line crossing portion. It uses an air-bridge. Therefore, at high frequencies such as microwaves, radio waves leak through the capacitance generated between the two lines at the intersection, and the isolation between the input and output lines deteriorates. As a result, there is a problem that an unwanted wave appears at the output terminal even when the switch provided between both lines performs an ideal operation. Also, since many such intersections are used in a normal switch matrix, as a result of leaked radio waves interfering with each other,
There is a problem that the performance of the entire switch matrix is significantly deteriorated. Further, in the microwave switch matrix forming the cross-point part of the conventional example 2, due to the structure, as the size of the matrix increases, the number of single circuits increases correspondingly, and the number of connecting parts increases, so that the assembly work is performed. Had the problem of being very troublesome.

The present invention has been made in order to solve the above problems, and in a microwave switch matrix having a large number of intersections, isolation between input / output terminals can be achieved without impairing reliability. It aims at improving and obtaining a switch without assembly work.

[0009]

A switch used in a microwave switch matrix or the like according to the first invention has at least two layers of a first layer and a second layer made of a dielectric material or a semiconductor. A ground conductor is formed between the two layers, and a first strip conductor (an example of a first feeding conductor) is provided on the first layer so as to face the ground conductor. And a second strip conductor (an example of a second feeding conductor) is provided in the second layer so as to face the ground conductor.
Via a semiconductor switch device having a switch function, which comprises a microstrip line, wherein at least one layer of the two layers is a semiconductor layer, and which uses a semiconductor element formed in the semiconductor layer. When connecting the first strip conductor and the second strip conductor, via holes (one example of connecting means) in the holes of the ground conductor are used for connection.

In the second aspect of the invention, for example, the thickness of the ground conductor is made twice or more the thickness of the skin current based on the power generated in the strip line and the required frequency.

[0011]

In the first invention, for example, a semiconductor layer-ground conductor-dielectric layer or a semiconductor layer-dielectric layer-ground conductor-dielectric layer sandwich structure is used as a semiconductor switch element on a semiconductor substrate. Since the line structure is integrally formed and two microstrip lines are independently configured using the respective layers, the coupling between the lines provided in the upper and lower layers can be sufficiently reduced. As a result, there is no leakage of radio waves at the input / output line intersection,
The isolation between the input and output terminals of the switch matrix is improved. Furthermore, since these configurations can be integrally formed by a semiconductor process, the accuracy of interlayer connection can be increased, and even a microwave switch matrix having a large number of intersections can be manufactured by a single semiconductor process, thus realizing a highly reliable switch. it can.

In the second aspect of the invention, the interference between the current flowing through the ground conductor by the input line and the current flowing through the ground conductor by the output line can be reduced, and as a result, good characteristics can be obtained in the microwave.

[0013]

EXAMPLES Example 1. Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is an external perspective view of a switch matrix according to an embodiment of the present invention, showing the intersections thereof. 2 is a perspective view showing the internal structure. In the figure, 7 is a semiconductor substrate, 9 is a field effect transistor, 12 is a first dielectric layer formed on the semiconductor substrate 7, and 13 is a second dielectric layer formed on the first dielectric layer 12. Body layer, 14 is a common ground conductor formed between the first dielectric layer 12 and the second dielectric layer 13, and 15 is a first dielectric layer so as to face the common ground conductor 14. A first strip conductor 12 provided on the second dielectric layer 13 is provided so as to face the common ground conductor 14.
The second strip conductor 18 provided in the common ground conductor 14 is a hole provided in the common ground conductor 14. The first dielectric layer 12, the common ground conductor 14 and the first strip conductor 15 constitute an output line 4d. Further, the second dielectric layer 13, the common ground conductor 14 and the second strip conductor 16 constitute an input line 2d. The gate of the field effect transistor 9 is connected to the input line 2d through the via hole 17a provided in the hole 18, and the drain is connected to the output line 4d. Further, the source is connected to the common ground conductor 14 via a via hole 17b. Here, a bias applying circuit is further required for the operation of the field effect transistor 9, but it is omitted here. The semiconductor substrate 7 has a thickness of 600
A high-purity GaAs substrate of about 100 μm is easily available. In addition, the first dielectric layer 12 and the second dielectric layer 1
For example, a nitride such as SiN, an oxide such as SiO2, an organic compound such as polyimide, etc. can be easily formed on the semiconductor substrate 7 by vapor phase growth or coating.

The radio wave propagating through the input line 2d is switched to the output line 4d side in accordance with the bias applied to the field effect transistor 9 in the same manner as in the conventional case. Here, since the input line 2d and the output line 4d have a structure shielded by the common ground conductor 14, the isolation between the input and output lines can be ensured, and there is a problem due to crosstalk or coupling. Does not happen. Therefore, in the switch matrix constituted by the lines having such a structure, there is an advantage that the isolation between the input and output terminals is improved and good performance is obtained. Furthermore, since these configurations are integrally formed by a semiconductor process, the accuracy of interlayer connection can be increased, and a highly reliable switch can be realized even in a microwave switch matrix having a large number of intersections.

Here, if the thickness of the common ground conductor 14 is set to be at least twice the thickness of the generated power and the skin current at the required frequency, the current flowing through the ground conductor of the input line 2d and the output line 4d. As a result of less interference with the current flowing through the ground conductor, good characteristics can be obtained in the microwave. 3 is a cross-sectional view taken along the line AA of FIG. 2, in which the arrow indicates an electric field, W is the thickness of the common ground conductor 14, W1 is the thickness of the skin current of the second strip conductor 16, and FIG. W2 is the thickness of the skin current formed on the common ground conductor 14 by the second strip conductor 16, W3 is the thickness of the skin current of the first strip conductor 15, and W4 is common by the first strip conductor 15. It is the thickness of the skin current formed on the ground conductor 14. As shown in this figure, if W> W2 + W4, the skin current interference is eliminated, and good characteristics can be obtained. For example, when W2 = W4 = 5000 angstroms, the thickness W of the common ground conductor 14 is 1
It is desirable that the thickness is 0000 angstrom (1 micron) or more, for example, 2 to 3 microns. Therefore, the semiconductor process for forming the common ground conductor 14 needs to have a step of ensuring a sufficient thickness of the common ground conductor.

Example 2. In the first embodiment described above, the field effect transistor having one gate is used as the switch for connecting the input and the output, but the present invention is not limited to this, and the dual gate having two gates is used. A field effect transistor may be used to further enhance the performance of the switch.

Embodiment 3. In the above-mentioned Example 1, the case where two dielectric layers were formed on the semiconductor substrate was shown. However, the present invention is not limited to this, and as shown in FIG. 4, the first dielectric layer 12 and the semiconductor layer 7 are formed on the dielectric substrate 13, and the field effect transistor 9 and the semiconductor layer 7 are formed.
It may be configured to be formed in. Further, the same effect can be obtained by replacing the first dielectric layer 12 and the semiconductor layer 7 in the structure shown in FIG. Further, the dielectric layer or the semiconductor layer may be composed of a plurality of dielectrics or semiconductors having different permittivities or thicknesses.

In Examples 1 to 3 described above, at least three layers of the first layer, the second layer, and the third layer made of a dielectric material or a semiconductor are provided, and at least one of the above three layers is provided. One layer is a semiconductor layer, and a single-sided ground conductor is formed between the first layer and the second layer, which are adjacent to each other among the three layers, and a first strip conductor is connected to the single-sided ground conductor. A semiconductor element provided on the first layer so as to face each other, and a second strip conductor provided on the second dielectric layer so as to face the one-side ground conductor and formed on the semiconductor layer. A circuit for connecting the first strip conductor and the second strip conductor to each other through a device having a switch function, and applying a bias voltage to the semiconductor element. Of each layer and connections between layers And means has been described microwave semiconductor switch matrix, characterized in that formed by integrally formed by a semiconductor process.

Example 4. In the above-mentioned Examples 1 to 3, the case where there are at least three layers is shown, but it may be two layers as shown in FIGS. 5 and 6.

[0020]

As described above, according to the first aspect of the present invention, a sandwich structure of a dielectric layer-ground conductor-dielectric layer, etc. and a via hole for interposing each layer are integrated with a semiconductor element on a semiconductor substrate. Since the microstrip lines are formed by using the respective dielectric layers, a switch matrix with good isolation between input and output terminals can be obtained without causing adverse effects due to line coupling between layers. realizable. Further, since these configurations can be integrally formed by a semiconductor process, the accuracy of interlayer connection can be increased, and a highly reliable switch can be realized even in a microwave switch matrix having a large number of intersections.

According to the second aspect of the invention, since the ground conductor is thickened, it is possible to obtain a good switch in which the skin current does not interfere.

[Brief description of drawings]

FIG. 1 is an external perspective view showing a main part of a microwave switch matrix according to an embodiment of the present invention.

FIG. 2 is an exploded perspective view showing an internal structure of a main part of a microwave switch matrix according to an embodiment of the present invention.

3 is a cross-sectional view taken along the line AA of FIG.

FIG. 4 is an exploded perspective view showing an internal structure of a main part of a microwave switch matrix according to another embodiment of the present invention.

FIG. 5 is an exploded perspective view showing an internal structure of a main part of a microwave switch matrix according to another embodiment of the present invention.

FIG. 6 is an exploded perspective view showing an internal structure of a main part of a microwave switch matrix according to another embodiment of the present invention.

FIG. 7 is a configuration diagram of a conventional microwave switch matrix.

FIG. 8 is an operation explanatory diagram of a conventional microwave switch matrix.

FIG. 9 is a perspective view showing a main part of a conventional microwave switch matrix.

FIG. 10 is a perspective view showing a main part of another conventional microwave switch matrix.

[Explanation of symbols]

 1 Input Terminal 2 Input Line 3 Output Terminal 4 Output Line 5 Switch 6 Termination Resistor 7 Semiconductor Substrate or Semiconductor Layer 8 Air-Bridge 9 Field Effect Transistor 10 Through Hole 11 Ground Conductor 12 First Dielectric Layer 13 Second Dielectric layer or dielectric substrate 14 common ground conductor 15 first strip conductor (an example of a first feeding conductor) 16 second strip conductor (an example of a second feeding conductor) 17 via hole (of connecting means) One example) 18 holes

Claims (2)

[Claims] 1. A switch in which the following elements are integrally formed by a semiconductor forming process: (a) first and second power feeding conductors; (b) a semiconductor layer having a semiconductor switch connecting the first power feeding conductor; (c) ) A ground conductor having a hole at a predetermined position between the semiconductor layer and the second feeding conductor, (d) a connecting means for connecting the semiconductor switch and the second feeding conductor through the hole of the ground conductor, 2. The switch according to claim 1, wherein a thickness of the ground conductor is equal to or larger than a thickness at which skin currents generated on both sides of the ground conductor by the first and second feeding conductors do not interfere with each other. The switch according to claim 1, wherein