JPH0775282B2 - Microwave automatic load matching circuit - Google Patents

Description

The present invention relates to a load matching circuit for stabilizing microwave power applied to a load in, for example, microwave heating.
In particular, it relates to a microwave automatic load matching circuit for automatically performing load matching.

[Conventional technology]

2. Description of the Related Art In recent years, the field of use of microwave power in various types of work is expanding more and more. In this case, it is important for quality control to always stabilize the microwave power applied to the load, and so-called load matching is performed for that purpose.

In the conventional additional matching in the microwave band, while monitoring the traveling wave power and reflected wave power to the load, manually operate three or more stub matching devices and E / H matching devices to minimize the reflected waves. The traveling wave power was adjusted to the maximum point.

However, in the manual load matching as described above, it is necessary to constantly monitor the load fluctuation and perform the corresponding operation without delay, and it is difficult to correctly and promptly respond to a sudden change or fluctuation.

Therefore, the present applicant has previously filed Japanese Patent Application No. 61-160362 (Japanese Patent Laid-Open No. 63-160362).
15502) proposed a microwave automatic load matching circuit capable of automatically performing the above load matching.

[Problems to be Solved by the Invention]

However, although the provisional purpose of automatically performing load matching by the microwave automatic load matching circuit has been achieved, there is a problem that the structure becomes complicated due to the use of the composite stub matching device.

The present invention has been proposed in view of the above problems, and has a simple structure, and can perform load matching quickly and automatically even when the load suddenly changes, and can always supply stable microwave power. An object is to provide an automatic load matching circuit.

[Means for Solving the Problems]

In order to achieve the above object, the microwave automatic load matching circuit according to the present invention has the following configuration.

That is, a groove is provided in the tube axis direction of a waveguide or a coaxial tube, one end of which is connected to the load circuit and the other end is connected to the signal source side. Through a single matching element made of metal or dielectric that is inserted into the pipe so as to be able to move forward and backward, and is further arranged on the same pedestal side closer to the signal source than the matching element, and the traveling wave component and reflected wave to the load circuit Provided with a plurality of signal detection probe to detect the component,
Based on the intensity and phase difference of the traveling wave component and the reflected wave component detected by the signal detection probe, the reflection coefficient | Γ | of the load circuit including the matching element and the cosine product of the argument θ |
The detection circuit outputs corresponding to Γ | cos θ and sine product | Γ | sin θ are obtained, and the load is automatically matched by changing the position of the gantry and the insertion length of the matching element into the tube.

[Action]

When the load is matched by using a single matching element as described above, for example, a single stub matching device, in the Smith diagram of FIG. 1, the admittance indicated by the load is the central (1,0) point. It is well known that on a passing g = 1 circle, only a single stub adjustment is needed to achieve perfect alignment. Therefore, g =
For the admittance outside the circumference of 1, if the distance between the load and the matching box is changed and adjusted so that it is on the circle of g = 1, a single stub matching box will completely match the load admittance over the entire area. Is possible.

If the matching stub is a metal rod, the insertion length is about 1 / the signal wavelength.
If it is 4 or less, it becomes capacitive (that is, the susceptance is positive), and beyond this point, it becomes inductive. Dielectric staps (bars or plates) exhibit capacitive properties.

In the admittance-Smith diagram of Fig. 1, at point B on the g = 1 circle, the standardized admittance is (1-jb), so insert a metallic stub at this point and set its susceptance to + jb.
Then, it becomes (1 + j0) and is matched. Also at point A-
Matching can be achieved by inserting the jb inductive stub.

Now, when the additional admittance is at point C on the circumference of g = 1, from this point in the direction of the signal source (Pa−Pc) λ (λ
Can be aligned by moving the stub by (tube wavelength) and inserting a metal rod stub. In the case of the derivative stub, the stub can be moved as much as (0.5−Pb + Pc) λ in the load direction to match the point B. That is, the single stub matching device may be moved to an appropriate position according to the reflection coefficient on the load side including the matching device, and a susceptance having an appropriate value may be inserted. At this time, the interval between the probe position of the detection circuit and the matching device is constant. That is, the detection probe and the matching device are fixed on the same mount and moved along the groove of the tube axis.

Specifically, the signal detection probe detects the traveling wave component and the reflected wave component flowing in the tube, and based on the intensity and phase difference between them, the reflection coefficient | Γ | of the load circuit including the matching device.
Then, the detection circuit output corresponding to the cosine product | Γ | cos θ and the sine product | Γ | sin θ of the deviation angle θ is obtained. The gantry is moved by this one output voltage, and the insertion length of the stub matching device is changed by the other output voltage to perform automatic matching.

Since the cosine product and the sine product are quadrature components, either one may be used for moving the gantry, but here, for the time being, the gantry is moved by the output voltage of the cosine product │Γ | cosθ and The description will be made assuming that the insertion length of the stub matching device is changed.

The signal detection probe should be as close to the stub matching device as possible,
Alternatively, it is preferable to shift the wavelength by about 1/2 of the guide wavelength. Now, assuming that the cosine product | Γ | cosθ shows a positive value, if the gantry is moved in the load direction, when the load admittance is in the quadrants I and IV, the gantry moves in the load direction and the quadrant II. And if it is in the third quadrant, it will move toward the signal source. The shaded area in FIG. 2 indicates the movement toward the signal source.

The capacitance of the metal rod is adopted as the stub matching device, and the sine product Γ
If the stub insertion length is changed by the output voltage of sin θ and the insertion length is made shallow when the value is positive, the insertion length is made deep when the value is negative. That is, as shown in FIG. 3, for reflection coefficients in the I and II quadrants, | Γ | sin θ has a positive value, so that the insertion length of the capacitive stub decreases and the admittance diagram shows the same conductance. Start moving downwards on the circle so that the capacitive susceptance decreases, III and IV
In the quadrant (in the shaded area in FIG. 3), the movement starts upward on the same g circle. However, since the equivalent conductance g also changes with the change of the phase angle θ, the matching point of the matching point (1 + j0) rapidly reaches with the change of the equivalent susceptance b.

For example, in the case of the reflection Γ ₁ of the normalized admittance (g ₁ + jb ₂ ) in the I quadrant on FIG. 4, the cradle moves in the load direction due to the positive output voltage of the cosine product, and the insertion length of the matching box is increased. Is reduced, so that g ₁ and b ₂ are both reduced to reach the matching point of (1, + j0).

In the case of the reflection coefficient Γ _{2 in} the _second quadrant II, | Γ | cos θ is negative, the gantry moves toward the signal source, and the matching device insertion length also becomes shallow, so g ₃ increases and b ₁ decreases Decrease and reach alignment.

In the case of Γ ₃ in the _third quadrant, the gantry moves toward the signal source and moves into the second quadrant, so that the alignment is performed in the above process.

Also, in the case of Γ ₄ in the IV quadrant, the gantry moves in the load direction and moves into the I quadrant, so that the alignment is performed in the same manner as above.

At this time, in the vicinity of the matching point, reciprocating motion may occur when the sensitivity is increased, but this can be solved by providing a dead zone in which the cradle movement and the slab insertion length do not change with a very small control voltage.

〔Example〕

FIG. 5 is an explanatory diagram of a structure when the present invention is applied to a waveguide circuit. At the center of the wide surface of the rectangular waveguide 1, a slit 2 is provided along the tube axis, and a stub matching device 4 and a signal detection probe 5 ₁ , 5 ₂ are mounted on a pedestal 3 that moves along the slit 2. There is.
The load circuit is connected to the right side of the figure, that is, to the matching device side, and a signal output is applied from the probe side.

The traveling wave component and reflected wave component of the signal flowing in the waveguide are extracted by the probe, and the cosine product |
The control voltage corresponds to Γ | cos θ and sine product | Γ | sin θ. And the electric motor 6 ₁ and 6 ₂ These output voltages to move the platform 3, respectively, by changing the insertion length of the stub matching device 4, to match the automatically load circuit.

As shown in FIG. 6, the load circuit 8 is connected to the matching device side waveguide port of the automatic matching device main body 7 of the present invention, and the signal source circuit 9 is connected to the other. Vc is a detection circuit output voltage corresponding to | Γ | cos θ, which is amplified by the amplifier 10 ₁ and then drives the electric motor M ₁ to move the gantry 3.

The voltage Vs corresponding to | Γ | sin θ is also amplified by 10 ₂ , and then the electric motor M ₂ is driven to change the insertion length of the stub matching device 4.

There are various methods for extracting the traveling wave component and the reflected wave component from the signal flowing in the waveguide and obtaining the cosine product and sine product of the reflection coefficient from them. In the present invention, the directional coupler method and the multi-probe method have been tried.

FIG. 7 shows a case where a directional coupler is used, and as the coupler, a so-called CM coupler can be used as well as a 1/4 wavelength coupler. Figure 5 ₁ traveling wave component extraction coupler, 5 ₂ is a reflected wave component extraction coupler. The output of the combiner 5 ₁ is divided into two by the signal divider 12, and the outputs thereof are applied to the synthesis detectors 14 ₁ and 14 ₂ , respectively. While the output of the reflected wave coupler 5 ₂ the components were detected by 90 degrees component divider 13 is divided into components with components and 90 degree phase difference between the input and phase,
It is added to the other input terminals of the synthetic detectors 14 ₁ and 14 ₂ . Each of these composite detectors square-detects the sum and difference voltages of the input signals, and the output of one of them is inverted in polarity.

Now, assuming that the angle of deviation of the reflection coefficient Γ of the load is θ, the traveling wave component output is a, and the reflected wave component output is b, the following equation is obtained.

In the combining detector 14 _1, by taking the difference of the squares of the voltage of the sum and difference of the two input voltages, the combined output voltage V _A is as follows.

V _A = 4ka ² | Γ | cos θ One of the combined detectors 14 ₂ has a reflected wave component output with a 90-degree phase difference, so the combined output voltage V _B is expressed by the following equation.

V _B = 4ka ² | Γ | sin θ Both V _A and V _B have a coefficient of 4ka ² , so if necessary, these are normalized to 1, so that the traveling wave component detecting coupler 5 ₁ use the output voltage, with normalization circuit 15 ₁ and 15 _2, V _a
And V _B are | Γ | cosθ and | Γ | sinθ, and | Γ | cos
The voltage V _c corresponding to θ and the voltage V _S corresponding to | Γ | sin θ
As a result, the circuit shown in FIG.

The multi-probe method used as an impedance direct-view device in the microwave band is also used as a cosine product / sine product generating circuit. There are a 4-probe method and a 5-probe method, but the 4-probe method is sufficient as the automatic load matching circuit of the present invention. For example, when the four-point probe method described in detail in "Microwave / millimeter wave measurement" by Bunichi Oguchi and Masamitsu Ota (Corona Corp., pp. 84-85) is applied to the present invention, as shown in FIG. Become.

The probe with detector 5 ₁ , 5 ₂ , 5 ₃ and 5 ₄ is arranged at the center of the wide surface of the rectangular waveguide and at a distance of 1/8 of the average in-tube wavelength. At this time, from the probe position closest to the load, the reflection coefficient viewed from the load side is | Γ | expθ, the detected voltage of the probe corresponding to the input voltage is Vi, and the detection characteristics are considered to be squared. When K is set as K, the output voltage of each probe is as follows.

V ₁ = K | Vi | ² (1+ | Γ ｜ ² +2 | Γ | cos θ) V ₂ = K | Vi | ² (1+ | Γ ｜ ² +2 | Γ | sin θ) V ₃ = K | Vi | ² (1+ │Γ│ ² +2 | Γ | cos θ) V ₄ = K | Vi | ² (1+ | Γ ｜ ² +2 | Γ | sin θ) Now, V ₁ and V ₃ are added to the input of the operational amplifier 11 ₁ and the difference is calculated. It can be found that V _A = V ₁ −V ₃ = 4K | Vi | ² | Γ | cosθ or the differential voltage between V ₄ and V ₂ with the operational amplifier 11 ₂ , V _B = V ₄ −V ₂ = 4K | Vi | ² | Γ | sin θ. If necessary, divide these V _A and V _B by a correction signal with a value corresponding to the traveling wave component to normalize, and correspond to | Γ | cos θ
The normalization circuits 15 ₁ and 15 _{2 set} the voltage V _s corresponding to V _c and | Γ | sin θ. The V _c and V _s are applied to the control circuit of FIG. 6 to perform automatic matching.

The standardization circuits 15 ₁ and 15 _{2 in} FIGS. 7 and 8 are used to reduce the change in the automatic matching sensitivity due to the power used, and there is often no need for that. The four-probe method is often adopted in the practice of the present invention.

〔The invention's effect〕

As described above, according to the present invention, a single matching element such as a single stub matching device enables an automatic load matching circuit over the entire area. Almost the same effect was obtained by using a single stub matching device instead of a complex compound stub matching device or E / H matching device. The response speed was also sufficient for use. For example frequency 2.
When a VSWR10 load was connected at 45 GHz, the matching circuit according to the present invention matched VSWR 1.1 or less within 1 second.

It should be noted that although the above embodiment has been described as a rectangular waveguide circuit, the present invention can be effectively used in a coaxial waveguide circuit, and similar effects can be obtained.

[Brief description of drawings]

FIG. 1 is an operation explanatory diagram of a single stub matching device, FIG. 2 is an explanatory diagram of movement of a mount of an automatic load matching circuit of the present invention, FIG. 3 is an explanatory diagram of the susceptance change, and FIG. 4 is a load matching device. Explanatory drawing, FIG. 5 is a schematic view of the matching device main body of the present invention, FIG. 6 is a drive circuit diagram by an electric motor, FIG. 7 is a control voltage detection circuit diagram by a directional coupler, and FIG. 8 is a 4-probe method. FIG. 6 is a control voltage detection circuit diagram when adopted. Reference numeral 1 is a grooved waveguide, 2 is a coupling groove, 3 is a movable mount, 4 is a stub matching device, 5 ₁ , 5 ₂ , 5, ₃ , 5, ₄ and 5 ₅ are detection probes, and 6 ₁ and 6 ₂ are electric motors. , 7 is a matcher body, 8 is a load circuit, 9 is a signal source circuit,
10 ₁・ 10 ₂ is an amplifier, 11 ₁・ 11 ₂ is a differential amplifier, 12 is a signal divider, 13 is a 90-degree component divider, 14 ₁・ 14 ₂ is a synthetic detector, 15 ₁・ 1
5 ₂ is a standardization circuit, 16 ₁・ 16 ₂・ 16 ₃・ 16 ₄ is a detector.

Claims (1)

[Claims] 1. A groove is provided in a pipe axis direction of a waveguide or a coaxial pipe, one end of which is connected to a load circuit and the other end is connected to a signal source side, and a pedestal movably provided along the groove is provided. A single matching element made of metal or dielectric that is inserted in the tube through the groove so as to be able to move forward and backward, and is further arranged on the same stand closer to the signal source side than the matching element, and is a traveling wave component to the load circuit. And a plurality of signal detection probes for detecting reflected wave components, and the reflection of the load circuit including the matching element based on the intensity and phase difference between the traveling wave component and the reflected wave component detected by the signal detection probes. A detection circuit output corresponding to the cosine product | Γ | cosθ and the sine product | Γ | sinθ of the coefficient | Γ | and its argument θ is obtained, and the position of the gantry and the insertion length of the matching element into the pipe are changed by these. Automatic load matching Black wave automatic load matching circuit.