JPS5912204B2 - microwave solid state oscillator - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a microwave solid-state oscillator using a semiconductor oscillation element, which is particularly easy to configure and adjust, and is designed to highly stabilize the oscillation frequency.

Invat diodes and Gunn diodes are mainly used as semiconductor oscillation devices in the microwave band.
Since these are semiconductor devices, they have large variations in characteristics and temperature dependence, and when configuring an oscillator using a microstrip line, it is extremely difficult to set the oscillation frequency due to the low no-load Q of the resonator. The temperature dependence of frequency also increases, and similar measures must be taken in applications that require a high degree of frequency stability.

Particularly when a Gunn diode is used as a local oscillator, frequency fluctuations due to temperature changes pose a serious problem.

As a method of solving such problems, there is a method of using a dielectric resonator that has good temperature stability and a high no-load Q.

FIG. 1 shows an example of a stabilized oscillator equipped with a bandpass P-wave device using a conventional Gunn diode.

Note that A in FIG. 1 is a top view, and B is a cross-sectional view taken along the line A--A.

In FIG. 1, a Gunn diode 3 is mounted on a microstrip line 2 constructed on a dielectric substrate 1 at a position as shown in the figure, and the oscillation frequency (hereinafter referred to as this) before stabilization is determined depending on the shape of the microstrip line 2. The original oscillation frequency is determined and stabilized by drawing the original oscillation frequency into the resonant frequency of the dielectric resonator 4, which is placed as a band-stop f-wave device at a position of 2π or approximately 2π in electrical angle from the Gunn diode 3. It is configured so that the output is taken out from the output terminal 5.

DC bias is supplied from terminal 6 through low pass filter 7.

In addition, 8 in the figure is a metal support that serves as a ground conductor for the microstrip line 2 and a heat sink for the Gunn diode 3.

In an oscillator with such a structure, the oscillation output and frequency of the oscillator before adding the dielectric resonator 4 (hereinafter referred to as the original oscillator) depend on the shape of the microstrip line 2 where the Gunn diode 3 is mounted. After adding the dielectric resonator 4 having almost the same resonant frequency as the original oscillation frequency, the oscillation frequency will be pulled into the resonant frequency of the dielectric resonator 4, and therefore, if the dielectric resonator 4 is stable, oscillation will occur. The frequency will also be stabilized.

However, in order to obtain the desired oscillation output and oscillation frequency with high stability in the above structure, it is necessary to optimize the oscillation output and oscillation frequency of the original oscillator excluding the dielectric resonator 4. The reality is that this has to be done through trial and error.

Even if optimization is done, it will be necessary to recreate the pattern when using diodes with different output levels.

Furthermore, the temperature stability of the frequency of the oscillator shown in FIG. 1 is ultimately determined by the stability of the dielectric resonator 4 if the fluctuation of the original oscillation frequency is within the pull-in range of the dielectric resonator 4.
When the pull-in range is exceeded, the oscillation frequency shifts to another oscillation frequency, and the stability deteriorates extremely.

Therefore, it is necessary to suppress fluctuations in the original oscillation frequency within the pull-in range of the dielectric resonator 4, but this cannot be achieved over a wide temperature range unless special temperature compensation is performed.

From the above, the conventional microwave solid state oscillator shown in FIG. 1 has the disadvantage that it is difficult to design and adjust and cannot be stabilized over a wide temperature range.

Furthermore, since the oscillator shown in FIG. 1 uses a dielectric resonator as a band-elimination type P-wave device, the oscillation characteristics change greatly in response to load fluctuations, and it also has the disadvantage that high frequency components of FM noise cannot be removed.

The present invention aims to eliminate such drawbacks of the conventional example, and an embodiment thereof will be described below with reference to FIG. 2.

Note that FIG. 2A is a top view, and FIG. 2B is a sectional view taken along the line BB.

In FIG. 2, a semiconductor oscillation element 9 is installed in contact with one edge of a dielectric substrate 10, and is electrically connected to one end of a microstrip line 12 connected to a branch terminal P of a hybrid circuit 11. DC bias is supplied from terminal 16 through low-pass filter 17 .

π/2 or almost π/ in electrical angle from the semiconductor oscillation element 9
2 distant points (reference plane T2) and hybrid circuit 1
π/in electrical angle from the open end (reference plane T6) of the microstrip line 15 with one end open connected to the branch terminal P3 of 1
2 or the first one at a location approximately π/2 apart. second 2
Two short cylindrical (or disc-shaped) TE (transverse electric) mode dielectric resonators 18 and 1
9 is arranged on the dielectric substrate 10 so as to be magnetically coupled to the microstrip lines 12 and 15.

The output is taken out from one end 22 of the microstrip line 13 connected to the branch terminal P2 of the hybrid circuit 11.

Further, in order to change the resonance frequency of the dielectric resonators 18 and 19, vertically movable disc-shaped metal plates 20 and 21 are placed facing each other above them.

Further, one end of the microstrip line 14 connected to the branch terminal P4 of the hybrid circuit 11 is open.

Note that 23 in the figure is a metal support that also serves as a ground conductor and a heat sink for the semiconductor oscillation element 9, similar to 8 in FIG.

Further, the hybrid circuit 11, the microstrip lines 12, 13, 14, 15, and the low-pass filter 17 can be continuously formed using known techniques such as printing and vapor deposition.

Next, the operation of this embodiment will be explained. Since the two dielectric resonators 18 and 19 have different functions, there is a case where only the dielectric resonator 18 is added, and a case where the two dielectric resonators 18 and 19 are added.
The case where both 8 and 19 are added will be explained separately.

First, an oscillator to which only the dielectric resonator 18 is added will be explained.

This oscillator corresponds to the original oscillator shown in FIG. 1 except for the dielectric resonator 4.

That is, while in FIG. 1 the original oscillation output and frequency are determined by the shape of the microstrip line 2, in this embodiment they are determined by the position and resonance frequency of the dielectric resonator 18 in FIG. are different.

At the resonant frequency of the dielectric resonator 18, the real part of the load impedance when looking at the load side from the coupling part (reference plane T2) of the dielectric resonator 18 and the microstrip line 12 is:
The equal economical conductance of the dielectric resonator 18 is the characteristic impedance Z of the microstrip line 12.

(usually chosen as 50Ω), so
It becomes larger than Zo, and the imaginary part of the load impedance becomes almost zero.

In addition, the dielectric resonator 18 and the microstrip line 12
By changing the distance between them, the degree of coupling changes and the magnitude of the equivalent conductance inserted in series with Zo can also be changed.

Furthermore, the real part of the impedance seen from the position of the semiconductor element (reference plane T1) which is electrically π/2 away from the reference plane T2 can be made smaller than the characteristic impedance 2° of the microstrip line 12 at the time of resonance.

That is, the load circuit viewed from the reference plane T1 can be operated as a single series resonant circuit with variable resistance.

Usually, a Gunn diode or an Imbat diode oscillates with a low impedance load, so by adjusting the position of the dielectric resonator 18 as described above, the real part of the load impedance seen from the semiconductor oscillation element 9 can be reduced. Oscillation output can be optimized.

At frequencies far from the resonant frequency of the dielectric resonator 18,
The load impedance is only the characteristic impedance of the microstrip line 12, and this value is usually 50Ω, which is larger than the load resistance that allows oscillation, and oscillation will stop.

That is, the load can be optimized only near the resonant frequency, and single mode oscillation without mode jumps can be achieved.

Furthermore, the oscillation frequency can be varied over a wide range by moving the metal plate 20 closer to or away from the dielectric resonator 18 to change the resonant frequency of the dielectric resonator 18.

Next, the effect of adding the second dielectric resonator 19 will be explained.

In order to stabilize the oscillation frequency, the load Q of the dielectric resonator 18 is increased so that the oscillation frequency is determined only by the resonant frequency of the dielectric resonator 18, and the temperature stability of the dielectric resonator 18 is improved. For example, although it is possible to obtain high stability without using the second dielectric resonator 19, it is difficult to reliably obtain high stability when the characteristics of the diodes vary widely.

This is because the unloaded Q of the dielectric resonator 18 cannot currently be increased to the point where the diode susceptance does not affect the oscillation frequency, so in order to improve temperature stability, the dielectric resonator 18 must have a positive temperature coefficient. In addition, it is necessary to adopt a method of canceling out the negative temperature coefficient based on the diode susceptance, because if there are variations in the diode susceptance, the temperature coefficient will not be canceled and the temperature stability will also vary.

Therefore, in order to eliminate this variation in temperature stability, the second dielectric resonator 19 is added.

The role of this second dielectric resonator 19 corresponds to the role of the dielectric resonator 4 in FIG.

That is, the desired oscillation output and oscillation frequency are first set using the dielectric resonator 18, and the oscillation frequency is finally determined by the resonant frequency of the dielectric resonator 19 by drawing the desired oscillation output and oscillation frequency to the resonant frequency of the dielectric resonator 19. It is.

Next, the role of the dielectric resonator 19 will be explained in more detail.

Appropriate reference planes T3 and T4 are provided for the branch terminals P1yP3 and P4 of the hybrid circuit 11, respectively.

T7 is set, the reference planes of the open ends of the microstrip lines 14 and 15 are respectively T8 and T6, and the coupling portion between the dielectric resonator 19 and the microstrip line 15 is defined as a reference plane T5.

The electrical angle between T2 and T3 is 01, the electrical angle between T4 and T5 is 02, the electrical angle between T7 and T8 is θ3, and the reflection coefficient of the dielectric resonator 19 is 17"cl<ejF.
c, the reflection coefficient at the reference surface T8 is l I” t l e
jvt and the oscillation output wave at the reference plane T2 is a,
The injection wave toward the diode on the 12th plane is expressed by the following equation (1).

Further, oscillation output P is taken out to branch terminal P2 of hybrid 11.

ut is given by the following equation (2). Here, η and ξ are angles given by the following equation.

At the resonant frequency of the dielectric resonator 19, it becomes.

However, Q 1 and Q 2 are the external Q and no-load Q of the dielectric resonator, respectively.

Since T8 is an open surface, It is.

Therefore, at resonance, η=-20-2θ2 −
・・・・・・・・・(6a) ξ=-2θ2-2θ3
......(6b).

Here, now θ + θ3 = g/2...
If you select [7] between songs, the output will be P.

ut is obtained from the above equation (2).

Moreover, at this time, the injection wave becomes the following equation.

It is well known that in the conventional example shown in FIG. 1, stabilization can be achieved by adjusting the position of the dielectric resonator and appropriately selecting the phase of the injected wave.

In the embodiment shown in FIG. 2 as well, as can be seen from equation (9) above, the phase of the injection wave can be controlled by appropriately selecting η and therefore θ1+02.

From this, it is clear that the embodiment shown in FIG. 2 can achieve the same stabilization as the conventional example shown in FIG.

Therefore, in the oscillator of the present invention, the first stage is highly stabilized by the dielectric resonator 18, and the dielectric resonator 19 has a high no-load Q and a temperature coefficient of zero or close to zero. By achieving even higher level of stabilization in the second stage, variations in the diodes can be absorbed and high stability can be reliably achieved.

In addition, since it is easy to bring the fluctuation of the original oscillation frequency into the stabilization frequency range (pull-in range) of the second stage by the first-stage stabilization, the oscillation frequency jumps (pull-in range) as in the conventional example. This is difficult to occur even if the operating temperature range is widened, and single mode oscillation can be achieved over a very wide temperature range.

Furthermore, in the conventional example shown in Figure 1, load fluctuations directly affect the load impedance seen from the diode to the dielectric resonator, so there is a drawback that the stabilization characteristics and oscillation output are easily affected by load fluctuations. However, since the present invention uses a hybrid circuit, it is clearly less susceptible to load fluctuations and is superior to the conventional example.

Furthermore, in the present invention, the maximum output is obtained at the resonant frequency of the dielectric resonator 19, as can be seen from the above equation (8), and the hybrid circuit including the dielectric resonator operates like a type of transmission type R-wave generator. Therefore, high frequency components of FM noise can also be reduced.

As described above, since the present invention performs stabilization in two stages, it is possible to achieve a very high degree of stability even when there are variations in the characteristics of the semiconductor oscillation elements used, and furthermore, it is possible to achieve a very high degree of stability. It can be produced safely within a temperature range.

Furthermore, since the present invention allows the load impedance seen from the semiconductor oscillation element to be adjusted by adjusting the position of the dielectric resonator, the oscillation output can be easily optimized, and when semiconductor oscillation elements with different characteristics are used. It also has a number of interesting features such as being able to optimize the oscillation characteristics without changing the microstrip line.

Further, the present invention is superior in that it does not have the drawbacks of the conventional band-stop E wave device loading type stabilization method, which are susceptible to load fluctuations and cannot remove high frequency components of FM noise.

In addition, although the above-mentioned embodiment of the present invention uses a short cylindrical or disc-shaped dielectric resonator, the shape of the resonator may be a rectangular parallelepiped or other dielectric resonator as long as it resonates in the TE mode. , any shape can be used.

[Brief explanation of the drawing]

FIGS. 1A and 1B are a top view and a sectional view taken along the line A-A of a conventional band-stopping P-wave device-equipped microwave solid-state oscillator, and FIGS. It is a sectional view taken along the B line. 9... Semiconductor oscillation element, 10... Dielectric substrate, 11... Hybrid circuit, 12, 1
3,14゜15・・・Microstrip line,
16...DC bias supply terminal, 17...
...Low pass filter, 18.19...Dielectric resonator, 20,21...Metal plate, 22...
...output terminal, 23...metal support stand.

Claims (1)

[Claims] 1. The first structure constructed on a dielectric substrate. Second. 3. A semiconductor oscillation element is connected to one end of the first branch microstrip line of the microstrip type hybrid circuit having a fourth branch microstrip line, and the semiconductor oscillation element is connected to the first branch microstrip line. π/2 along the electrical angle or π/2 apart from the first
A first TE mode dielectric resonator is installed on a dielectric substrate at a position close to the branch microstrip line, and a second branch microstrip line serves as a main line for the first branch microstrip line. A third branch microstrip line serving as a sub line to the first branch microstrip line is open at one end, and the third branch microstrip line is connected to the third branch microstrip line from this open end. π/ in electrical angle along the track
A second TE mode dielectric resonator is installed on a dielectric substrate at a position close to the third branch microstrip line, which is approximately π/2 away from the fourth branch microstrip line of the hybrid circuit. A microwave solid-state oscillator characterized by having a strip line configured so that one end is electrically open.