JP4655257B2 - 2-terminal pair isolator - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to the field of high-frequency isolators, particularly two-terminal pair isolators that have wideband characteristics of reverse loss and insertion loss.
[0002]
[Prior art]
As a current technical situation of high-frequency isolators, one terminal of a three-terminal pair junction circulator is generally terminated with a matching impedance. This junction type circulator is classified into two types, that is, a distributed constant type circulator and a lumped constant type circulator. The circulator has an irreversible electrical characteristic, and its structure is based on a structure in which a magnetic field is applied perpendicularly to the ferrite thin plate and a conductor is close to the center of the ferrite thin plate. The former distributed constant type is selectively used when the size of the ferrite thin plate is ¼ or more of the wavelength of the high frequency transmitted through the handled ferrite thin plate, and the latter lumped constant type is used when the size is 1/8 or less. As a matter of course, the lumped constant type is more suitable for downsizing.
[0003]
FIG. 8 shows a schematic structure diagram and a schematic circuit diagram of an isolator that is currently used in a mobile phone or the like using the principle of a three-terminal pair lumped constant type circulator. The ferrite thin plate G is made of garnet-type ferrite, and three central conductors L1, L2, and L3 are arranged on the upper surface at an angular interval of 120 degrees. One end of each center conductor is an input / output line of the terminal pair (1), (2), and (3), and the other end is connected to a common portion GR that is a ground conductor. Matching capacitors C1, C2, and C3 are connected in parallel between one end of each of the central conductors L1, L2, and L3 and the common portion GR. In order to realize an isolator, an energy absorbing resistor element R is attached between the terminal pair (3) and the common part GR. A permanent magnet is loaded so that a static magnetic field substantially perpendicular to the main surface of the ferrite thin plate G is applied, but it is omitted in the drawing. The structure shown in FIG. 8 is desired by carefully adjusting the direction and strength of the static magnetic field, the shape and size of the center conductors L1, L2, and L3, and the capacitance of the three matching capacitors C1, C2, and C3. Operates as a circulator at a frequency fo (hereinafter referred to as a center frequency). That is, the high frequency input from the terminal pair (1) is transmitted to the terminal pair (2), and the high frequency input from the terminal pair (2) is transmitted to the terminal pair (3) with little loss. It is not transmitted in the opposite direction. When the resistance element R is connected to the terminal pair {circle around (3)}, most energy is absorbed there, and a state is realized in which almost no high frequency propagates from the terminal pair {circle around (2)} to the terminal pair {circle around (1)}. That is, it is possible to realize an isolator which is an element that helps propagation in one direction and blocks it in the opposite direction.
The isolator having the structure shown in FIG. 8 has the advantage that the insertion loss is small and the bandwidth is wide. However, the isolator has the disadvantages that the number of parts is large and there is a limit to the reduction in thickness and size. .
[0004]
As another example, there are two-terminal pair lumped constant type isolators (Japanese Patent Laid-Open Nos. 52-134349 and 53-129561) as shown in FIG. This two-terminal-pair lumped constant type isolator requires only two center conductors L1 and L2 that are orthogonal to each other, and the matching capacitors need only be C1 and C2, and has a very simple structure. In addition, since the two central conductors are orthogonal to each other in this structure, even if the correct isolator operation is not established away from the vicinity of the center frequency (this will be referred to as in-band), that is, out of the band. It has been pointed out that there is an advantage that high attenuation can be obtained. A major feature of this structure is that the two terminals of the energy absorbing resistance element R are connected to one ends of the central conductors L1 and L2, respectively. The other ends of the center conductors L1 and L2 are connected to a common portion GR that is a ground conductor. As can be seen from the structure shown in FIG. 8, this structure is characterized by one central conductor and one matching capacitor, which is extremely convenient for miniaturization and thinning of high-frequency isolators. It is.
[0005]
[Problems to be solved by the invention]
However, the structure of the two-terminal-pair lumped constant isolator shown in FIG. 7 has not been put into practical use until now. The reason is that although the bandwidth of reverse loss is certainly wide, it has the disadvantage that the bandwidth of insertion loss is narrow. Accordingly, the value of the insertion loss itself is not so small as compared with the structure of the three-terminal pair isolator in FIG.
As one example, when trying to widen the bandwidth, it is conceivable to reduce the normalized operating magnetic field σ (described later) by weakening the static magnetic field. However, this increases the magnetic loss, resulting in an increase in insertion loss. Another reason is that the operating principle has not been studied so finely as in the case of a three-terminal pair circulator.
[0006]
Thus, the inventors of the present invention have developed an original circuit simulator capable of analyzing the circuit of the two-terminal pair lumped constant isolator shown in FIG. 7, and have obtained various basic knowledge based on the circuit simulator. The operation principle of the two-terminal pair lumped constant isolator shown in FIG. 7 is briefly described below based on circuit analysis.
The high frequency entered from the terminal pair (1) causes a current to flow through the central conductor L1 and excites the ferrite thin plate G. The ferrite thin plate G is a permanent magnet that is magnetized in the direction of its main surface, and generates a high-frequency magnetic field component that is coupled to the orthogonal central conductor L2. This is due to the ferromagnetic resonance effect of ferrite in the microwave band. Without this effect, energy will not propagate to the center conductor L2. The matching capacitors C1 and C2 constitute a parallel resonance circuit that resonates at the center frequency fo in pairs with the center conductors L1 and L2. What should be noted here is a phase change in the case of propagation. That is, when energy is transmitted from the terminal pair (1) to the terminal pair (2), the phase difference between the input and output is 0 degree, and if the amplitude is the same at the input and output, no current flows through the resistance element R. . Conversely, when energy is transmitted from the terminal pair (2) to the terminal pair (1), the phase difference is exactly 180 degrees. At this time, a large current flows through the resistance element R, and energy is consumed by the resistance element. That is, energy is not easily transmitted from the terminal pair (2) to the terminal pair (1).
[0007]
However, this phenomenon is an ideal case. In practice, there is a line capacitance between the first and second central conductors, and there is a parasitic inductance in series with the resistance element. When such a parasitic element exists, the operation described above cannot be expected. Therefore, the present inventors have changed the line capacitance or the line capacitance when the intersection angle φ formed by the intersection of the center axis of the first center conductor L1 and the center axis of the second center conductor L2 is changed. A method to compensate for the parasitic inductance was investigated using an originally developed circuit simulator. This case has already been discussed in US patent 4,210,886. However, the theoretical background is unclear, and the conclusion reached has not necessarily calculated the optimum angle with practicality taken into account.
Therefore, the present invention reexamines the circuit characteristics of the two-terminal pair lumped constant type isolator in detail by using a uniquely developed circuit simulator, thereby obtaining an intersection angle φ between the first center conductor L1 and the second center conductor L2, Furthermore, the optimum condition of the attached circuit (for example, Cw in the case of the circuit of FIG. 4) connected between the first center conductor L1 and the second center conductor L2 is obtained, the structure is simple, and the size and thickness are reduced. In addition to the above, it is an object of the present invention to provide a very practically effective two-terminal pair isolator having a low insertion loss and a wide insertion loss bandwidth.
[0008]
[Means for Solving the Problems]
The present invention is a two-terminal pair isolator arranged near the center of a ferrite to which a static magnetic field is applied so that the first and second central conductors intersect each other in an electrically insulated state. One end of the second central conductor serves as first and second input / output terminals, the other end is connected to a common portion, and a first matching capacitor is provided between the first input / output terminal and the common portion. Is connected, a second matching capacitor is connected between the second input / output terminal and the common terminal, a resistance element is connected between the first and second input / output terminals, and the first This is a two-terminal pair isolator in which the crossing angle between the central axis of the central conductor of the second conductor and the central axis of the second central conductor is in the range of 40 to 80 degrees.
In the present invention, it is preferable to connect a third capacitor in parallel with the resistance element. The capacitance of the third capacitor is smaller than the capacitances of the first matching capacitor and the second matching capacitor.
In the present invention, it is also preferable to connect an inductor in parallel with the resistance element. In the present invention, an inductor is preferably connected in series with the resistance element. The common portion of the two-terminal pair isolator according to the present invention is connected to a ground conductor.
[0009]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. FIG. 4 is an equivalent circuit diagram of a two-terminal pair isolator which is the basis of the present invention. 7 differs from the conventional two-terminal pair isolator shown in FIG. 7 in that the crossing angle of the central conductor can be changed first and second, and in order to compensate for the effect of the crossing angle φ, a resistance element R is newly added. The third capacitor Cw is connected in parallel.
1 (a), (b), and (c) use a circuit simulator that was originally developed on the assumption that the center frequency fo is 1000 [MHz] and the equivalent circuit of FIG. 4 is an ideal circuit. The frequency characteristic of the S parameter of the two-terminal pair isolator calculated by The ideal circuit here means that the coupling between the two central conductors L1, L2 is perfect. In the figure, calculation results for three typical angles φ = 60 [degrees], φ = 90 [degrees], and φ = 120 [degrees] are shown. The parameters used at this time were characteristic impedance Zo = 50 [Ω], air core inductance K = 1 [nH], saturation magnetization of ferrite thin plate 4πMs = 900 [G], and resistance value of resistance element R = 50 [Ω ]. The calculated frequency range was set to 900 [MHz] to 1100 [MHz] of ± 10% with respect to the center frequency fo.
[0010]
The third capacitor Cw, which is an attached circuit element, is naturally Cw = 0 when φ = 90 degrees in the prior art. When φ = 60 [degrees], Cw = 7.85 [pF], and when φ = 120 [degrees], Cw = −7.85 [pF]. A negative Cw means that this part is not a capacitor but an inductor. This point will be described in detail later.
[0011]
FIG. 1A shows the frequency characteristic of the reflection loss S11. As can be seen from this figure, with φ = 90 [degrees] as a reference, when φ is smaller than this, the bandwidth of the reflection loss S11 is widened, and when φ is larger than this, the bandwidth is rapidly narrowed. Here, a value obtained by dividing the frequency width at which S11 falls below 20 [dB] by the center frequency fo is represented by a specific bandwidth W (S11) [%].
FIG. 1B shows the frequency characteristics of the insertion loss S21. As can be seen from this figure, on the basis of φ = 90 [degrees], when φ is smaller than this, the bandwidth of S21 is widened, and when φ is larger than this, the bandwidth is rapidly narrowed. Here, assuming that the value of the reflection loss S11 at 900 [MHz] is related to the bandwidth of the insertion loss, it is defined as IL [dB] (at 0.9fo). Indicated by Δ in the figure. A small IL corresponds to a wide bandwidth for insertion loss, and vice versa.
As described above, in FIGS. 1 (a) and 1 (b), the result that the bandwidth of reflection loss and insertion loss is widened at φ = 60 degrees is the focus of the present invention.
FIG.1 (c) shows the frequency characteristic of the reverse loss calculated on the same conditions. When φ = 90 [degrees], it shows a high reverse loss characteristic of 45 [dB] or more in the frequency range of 0.9fo to 1.1fo (900MHz to 1100MHz) as shown, but φ is larger than this Even if it becomes smaller, the reverse loss deteriorates. In particular, when φ is smaller than φ = 90 [degrees], the degree of deterioration increases. Here, an amount IS [dB] (at 0.96 fo) related to the bandwidth of reverse loss is defined. That is, this is the reverse loss value at 0.96fo (960 MHz in the figure). Indicated by Δ in the figure. A large IS corresponds to a wide reverse loss bandwidth, and a small IS corresponds to a narrow bandwidth.
[0012]
FIG. 2 shows the change of each attached parameter when the crossing angle φ of the two central conductors is changed in a wide range of 40 [degrees] to 140 [degrees] in order to show the principle and range of the present invention. . Here, the first matching capacitor and the second matching capacitor have the same value, and are represented by C (= C1 = C2). When φ is smaller than 90 [degrees], the third capacitor Cw gradually increases. When just φ = 60 [degrees], the third capacitor Cw, the first matching capacitor, and the second matching capacitor C match, and both are 7.85 [pF]. C tends to decrease, but is almost constant below 60 [deg.]. The resistance element R is constant at 50 [Ω].
Conversely, when φ is greater than 90 degrees, the capacitance of the first and second matching capacitors C increases rapidly as φ increases. On the other hand, the third capacitor Cw becomes negative, and its absolute value increases as φ increases. This is indicated by a fine dotted line in the figure. The curve of the absolute value of Cw is symmetrical with respect to φ = 90 [degrees]. Since there is actually no capacitor with a negative capacity, this can be equivalently represented as an inductor Lp. An equivalent circuit in this case is shown in FIG.
That is, when φ is greater than 90 degrees, an inductor Lp is required in parallel with the resistance element R. However, this equivalent circuit is not very practical. This is because, in practice, the behavior around 90 [degrees] is important, so I would like to discuss this part in detail, but at φ = 90 [degrees], Lp becomes infinite. In order to avoid this, a circuit in which an inductor Ls is inserted in series with a resistance element Rs as shown in FIG. 6 is practical. This is because when φ is brought close to 90 [degrees] from a large state, it gradually approaches Ls = 0 [nH] and Rs = 50 [Ω], so that there is continuity.
[0013]
The graph on the right half of FIG. 2 shows changes in Ls and Rs when φ> 90 [degrees]. As φ increases, Rs suddenly approaches zero, but Ls takes a maximum value of 115 degrees. When the angle becomes larger than that, Ls gradually decreases monotonously.
FIG. 3 shows the angular dependence of isolator characteristics calculated under the above conditions. IL (at 0.9fo) indicating the bandwidth of the insertion loss S11 decreases when φ is smaller than 90 [degrees], and takes a minimum value when φ = 60 [degrees]. On the contrary, when it becomes larger than 90 [degree], IL (at 0.9fo) rapidly increases. In addition, the normalized operating magnetic field σ indicating the strength of the static magnetic field is minimum at 90 [degrees], and becomes larger when the angle is larger or smaller. Here, the normalized operating magnetic field σ is a dimensionless number obtained by dividing the internal magnetic field Hi of the ferrite thin plate by the ferromagnetic resonance magnetic field Hres (= 2πfo / γ) at the center frequency fo. γ is a constant called magnetic rotation rate. The bandwidth W (S11) at which the reflection loss S11 falls below 20 [dB] increases as φ decreases, and takes a maximum value of 7.6 [%] at approximately φ = 60 [degrees]. When φ becomes larger than 90 [degrees], W (S11) decreases monotonously.
[0014]
On the other hand, IS (at 0.96 fo) indicating the bandwidth of the reverse loss shows a maximum value of 55 dB at φ = 90 [degrees], and decreases at both larger and smaller angles. In particular, it decreases monotonously when φ <90 [degree], and drops below 10 [dB] when φ = 40 [degree]. On the other hand, when φ> 90 [degrees], although it decreases, it shows a high attenuation of about 30 [dB].
Now, the practically effective results from FIGS. 3 and 4 are summarized as follows.
(1) When low insertion loss is important, a range of φ <90 [degrees] is desirable.
(2) When the reverse loss is important, φ = 90 [degree] is desirable.
(3) The insertion loss bandwidth and the reflection loss bandwidth are the largest at about φ = 60 degrees.
(4) Actually acceptable IS (at 0.96fo) = 10 [dB] is cut when φ is 40 degrees or less.
[0015]
Next, the idea of defining the claims of the present invention from the above results will be described.
That is, the present inventor theoretically clarified for the first time that the insertion loss bandwidth is extremely wide and the reverse loss is sufficiently practical at φ = 60 degrees. The basic idea of the present invention is not limited to φ = 60 degrees. That is, the gist of the present invention does not change even when IS (at 0.96 fo) indicating the bandwidth of the reverse loss is 40 degrees which is less than 10 dB. However, if φ becomes smaller than this, IS (at 0.96fo) becomes too small to be practically usable. For this reason, φ = 40 [degrees] was defined as the lower limit of the claims of the present invention. In addition, at φ = 80 [degrees], the insertion loss bandwidth IL (at 0.9fo) and the reflection loss bandwidth W (S11) are significantly improved compared to the prior art φ = 90 [degrees]. . Therefore, the gist of the present invention is fully alive even at φ = 80 degrees. However, if φ becomes larger than this, IS (at 0.96fo) increases, and the difference from the prior art becomes unclear. Therefore, φ = 80 [degrees] was defined as the upper limit of the present invention.
[0016]
Now, the third capacitor Cw is clearly described in the equivalent circuit of FIG. 4 which is the basis of the present invention. Here, the third capacitor Cw requires that the capacitance of Cw is considerably larger than C when the crossing accuracy φ = 40 [degrees] of the first and second central conductors, and φ = 80 [degrees]. Cw can be quite small.
Also, this Cw is not always necessary. This is because the two central conductors L1 and L2 intersect each other in the vicinity of the center of the ferrite thin plate G, and are electrically insulated by a thin insulating sheet. There is actually a line capacitance between them. This is because the line capacitance operates in exactly the same way as Cw in FIG. By skillfully utilizing this line capacitance, the effect of the present invention can be realized without the third capacitor Cw.
Due to the presence of the line capacitance, the third capacitor Cw is often practically smaller than the first and second capacitors C.
Also, this line-to-line capacitance may be too large and exceed the amount of Cw necessary to compensate for the effect of angle φ. In order to compensate for this overshoot, an inductor Lp may be connected in parallel with the resistance element R.
The circuit of the resistor element R and the inductor Lp connected in parallel thereto can be replaced by the resistor element Rs and the series inductor Ls.
[0017]
【The invention's effect】
As is clear from the above description, according to the two-terminal pair isolator of the present invention, the structure is simple with only the two central conductors, and not only is the size and thickness reduced, but also the insertion loss bandwidth A low-loss two-terminal pair isolator having a wide band can be provided.
[Brief description of the drawings]
1A is a frequency characteristic diagram of reflection loss of a two-terminal pair isolator showing the effect of the present invention, FIG. 1B is a frequency characteristic diagram of insertion loss of a two-terminal pair isolator illustrating the effect of the present invention, and FIG. The frequency characteristic figure of the reverse loss of the 2 terminal pair isolator which shows the effect of invention.
FIG. 2 is an angle dependency of an attached circuit parameter of a two-terminal pair isolator for realizing the effect of the present invention.
FIG. 3 is an angle dependency of two-terminal-pair isolator characteristics for explaining the effect of the present invention.
FIG. 4 is an equivalent circuit of a two-terminal pair isolator which is a subject of the present invention.
FIG. 5 is an equivalent circuit of a two-terminal pair isolator illustrating the effect of the present invention.
FIG. 6 is an equivalent circuit of a two-terminal pair isolator for explaining the effect of the present invention.
FIG. 7 is an equivalent circuit of a conventional two-terminal pair isolator.
FIG. 8 is an equivalent circuit of an isolator based on a prior art three-terminal pair circulator.
[Explanation of symbols]
▲ 1 ▼ ▲ 2 ▼ ▲ 3 ▼ I / O terminal pair
C1, C2, C3 matching capacitors
G Ferrite sheet
L1, L2, L3 Center conductor
GR Ground conductor
R resistance element φ Intersection angle of first center conductor and second center conductor
Cw parallel capacitor

Claims (2)

A two-terminal pair isolator arranged near the center of a ferrite to which a static magnetic field is applied so that the first and second central conductors intersect each other in an electrically insulated state;
One end of each of the first and second center conductors is a first and second input / output terminal, and the other end is connected to a common portion serving as a ground conductor, and between the first input / output terminal and the common portion. Is connected with a first matching capacitor, a second matching capacitor is connected between the second input / output terminal and the common portion , and a resistor is connected between the first and second input / output terminals. element is connected, Ri angle of intersection near the range of 80 degrees from 40 degrees with the central axis of said first central conductor center axis and the second center conductor,
A third capacitor connected in parallel with the resistive element;
The terminal pair isolator , wherein the capacitance of the third capacitor is smaller than the capacitance of the first matching capacitor and the second matching capacitor . A two-terminal pair isolator arranged near the center of a ferrite to which a static magnetic field is applied so that the first and second central conductors intersect each other in an electrically insulated state;
One end of each of the first and second center conductors is a first and second input / output terminal, and the other end is connected to a common portion serving as a ground conductor, and between the first input / output terminal and the common portion. Is connected with a first matching capacitor, a second matching capacitor is connected between the second input / output terminal and the common portion, and a resistor is connected between the first and second input / output terminals. Elements are connected, and an intersection angle between the central axis of the first central conductor and the central axis of the second central conductor is in the range of 40 degrees to 80 degrees,
A two-terminal pair isolator, wherein an inductor is connected in series with the resistance element.

Images