JPS624009B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides a magnetic amplifier that reduces costs by increasing the power efficiency of an amplifier and providing the amplifier itself with a power supply isolation function and a step-up effect on the power supply voltage. The purpose is to obtain an amplifier.

Currently, most low-frequency power amplifiers are class A or class B amplifiers using transistors or the like.
However, since the power amplification element itself consumes power in these amplifiers, the efficiency thereof is low, and there are problems in that the size, weight, cost, etc. for heat dissipation increase.

In addition, low-frequency power amplifiers usually have a configuration in which the commercial AC power supply is isolated with a transformer, the secondary output is rectified, and the DC voltage is supplied to the amplifier, or the DC voltage is supplied directly from a battery to the amplifier. In many cases, the structure is as follows.

However, with the isolation method using a transformer, there is a problem that the weight and cost of the transformer account for a large amount of weight, which becomes an obstacle to cost reduction.In addition, in the case of battery drive, the output voltage of the amplifier depends on the voltage of the battery. The problem is that there is a limit and no further output can be obtained.

In order to solve these problems, the present invention provides a magnetic amplifier that uses a transformer as an amplifier and allows the amplifier itself to have a power isolation function and a step-up effect on the power supply voltage. It is.

FIG. 1 shows an embodiment of the present invention. In Fig. 1, 1 is a square wave voltage generator, 2 to 4 are transformers, 5 to 7 are primary windings, 8 and 9 are secondary windings, 1
0 to 12 are control windings, 13 and 14 are bias windings, 15 and 16 are bias current supply circuits, 17 is a control amplifier, 18 is an input terminal, 19 to 22 are rectifying diodes, 23 and 24 are smoothing Capacitors, 25 and 26 are resistors that combine rectified output, 27
28 and 29 are feedback circuits, 30 is an output terminal, and 31 is a load.

In the above configuration, first, the transformer 2 is biased near the saturation region in the negative direction by the bias current of the bias winding 13, and the transformer 3 is biased near the saturation region in the positive direction by the bias current of the bias winding 14. ing. The transformer 4 is then unbiased and in the active region.

The operating state of each transformer is shown in FIG. 3a.
In FIG. 3a, the vertical axis is the magnetic flux density, the horizontal axis is the magnetic field strength, and a1, a2, and a3 represent the BH curves of the transformers 2, 3, and 4, respectively. Furthermore, P, Q, and R represent the operating points (bias points) of the transformers 2, 3, and 4 when no signal is present, respectively. When the operating point of the transformer is near the center of the BH curve, the inductance of the primary winding increases, and as it approaches saturation, the inductance of the primary winding decreases.

The relationship between magnetic field strength and inductance for each transformer is shown in Figure 3b. In Fig. 3b, b1, b2, b3 are transformers 2, 3,
S represents the inductance of the primary winding of transformer 4, S represents the operating point of transformers 2 and 3 when no signal is present, and T represents the operating point of transformer 4 when no signal is present.

As can be seen from FIG. 3b, when there is no signal, the inductance of the primary winding of transformer 4 is the largest, and the inductance of the primary windings of transformers 2 and 3 are both extremely small. For this reason, the square wave voltage of the square wave generator 1 mostly appears in the primary winding 7 of the transformer 4, and hardly appears in the primary windings 5, 6 of the transformers 2, 3. Therefore, almost no square wave voltage appears in the secondary windings 8, 9 of the transformers 2, 3, and their rectified outputs are also almost zero.

Next, when a positive control current is supplied from the control amplifier 17 to the control windings of the transformers 2, 3, and 4,
The operating points S and T in FIG. 3b move to the right. Therefore, the inductance b1 of the primary winding of the transformer 2 becomes larger, the inductance b2 of the primary winding of the transformer 3 becomes smaller and smaller, and the inductance b3 of the primary winding of the transformer 4 becomes smaller. Then, since the square wave voltage is distributed to the primary windings of transformers 2 and 4 according to the ratio of b1 and b3, the primary winding 5 of transformer 2 has c1 in FIG.
A voltage as shown in appears. Conversely, when a negative control current is supplied, a voltage as shown at c2 in FIG. 3c similarly appears in the primary winding 6 of the transformer 3. Also, the secondary windings 8 and 9 of the transformers 2 and 3 are
1, a voltage similar to c2 appears.

Therefore, when the square wave voltage appearing in the secondary windings 8 and 9 is rectified into positive and negative polarities by the rectifiers 19 to 22, rectified outputs as shown at c1 and c3 in FIG. 3c are obtained. And these rectified outputs are connected to resistor 2
When 5 and 26 are combined, positive and negative output voltages are obtained for positive and negative control currents as shown at c4 in FIG. 3c.

The voltage of transformers 2 and 3 is almost zero when there is no signal, and voltage is supplied from transformer 2 for positive output, and voltage is supplied from transformer 3 for negative output, so it is normal class B. Similar to the operation of an amplifier. Therefore, like a class B amplifier, the crossover distortion changes depending on how the positive side output and the negative side output overlap, and there is an optimal bias state where the distortion is minimized. The amount of overlap is bias winding 1
This can be adjusted by adjusting the currents flowing through 3 and 14. Transformers 2 and 3 usually have similar specifications. Therefore, bias winding 1
The currents 3 and 14 may be the same amount, and one may be supplied with the same polarity and the other with the opposite polarity with respect to the polarity of the control winding. An effective method for this purpose is to connect bias windings 13 and 14 in series with opposite polarities and supply bias currents from the same bias current supply circuits 15 and 16, as shown in FIG. In this case, for example, it is advantageous to configure the resistor 16 with a variable resistor because the bias amounts of the transformers 2 and 3 can be adjusted simultaneously.

By providing the weight as described above, an idling current flows between the positive and negative outputs of the rectifier circuit even when there is no signal. This current is limited by the resistors 25 and 26, but if the winding resistance of the transformers 2 and 3 and the resistance of the diodes 19 to 22 can be expected, the resistors 25 and 26 can be omitted and the positive and negative rectifier circuits It is also possible to directly connect the output to obtain the output.

In the example shown in Figure 1, double-wave rectifier circuits 19 to 22 are used as the positive and negative rectifier circuits, but by using double-wave rectification in this way, the ripple of the square wave component can be made very small. can do. Furthermore, in the example shown in FIG. 1, a ripple removal capacitor 27 is used between the output and ground in order to further suppress ripple components.

In the magnetic amplifier according to the present invention, a certain degree of linearity can be obtained in the power amplifying section itself using a transformer, but in order to further improve the linearity, it is necessary to
It is effective to apply negative feedback as shown in 8 and 29.

In that case, the transfer characteristics from the control windings of transformers 2, 3, and 4 to the primary or secondary windings will be delayed in phase due to the presence of leakage inductance, so
It is difficult to apply stable negative feedback over the high range, and the effect of improving linearity is reduced accordingly. Therefore, if the phase delay of the transformer is corrected by performing phase advance correction in the control amplifier, negative feedback can be applied to the high frequency range, and linearity can be improved accordingly. .

By the way, in the conventionally known magnetic amplifier,
Normally, a commercial power supply frequency (50, 60 Hz) is used as the AC voltage source to be controlled, and the power is controlled by a control signal of a lower frequency. However, in order to amplify low frequencies (for example, audio frequencies), which is the object of the present invention, it is not possible to use such conventional commercial power frequency power as is.

Therefore, in the present invention, a method is used in which a square wave having a frequency at least twice as high as the signal frequency to be handled is controlled. This square wave can be obtained from a square wave oscillator powered by DC voltage obtained by rectifying the commercial power supply or DC voltage from a battery, and using a square wave in this way can be generated with very little power loss. Therefore, it is compatible with one of the objectives of the present invention, which is high efficiency.

Also, in the embodiment shown in Fig. 1, three transformers are used, and the outputs of two of them are rectified and combined to provide the amplifier output, but as shown in the embodiment shown in Fig. 2, two transformers are used. Basically, the same functionality can be obtained in both cases.

In this case, when the input signal is zero, the voltage of the square wave voltage generator 1 is divided into two by the primary windings 5 and 6, so the rectifiers 19, 20 and 21, 22
1/2 of the maximum rectified output is generated at the output of the , and these are combined by resistors 25 and 26 to obtain zero output voltage. Power is consumed by the low resistors 25 and 26, which deteriorates the efficiency accordingly.As for the maximum output voltage, there is a large voltage drop in the resistors 25 and 26, so there is a problem that less than half of the maximum rectified output can be obtained. However, the basic function of amplifying low frequencies such as audio frequencies can be fully realized.

Next, the number of turns of the control winding will be explained. Regarding the number of turns of the control windings 10, 11, and 12, it is advantageous to have a large number of turns in order to reduce the control power, but the response characteristics deteriorate and, in particular, there is a limit to the maximum output voltage of the control amplifier. Therefore, there is a problem that the maximum voltage change speed (slew rate) of the output deteriorates, and conversely, if the number of turns is small, the slew rate improves, but there is a problem that the control power increases, and the optimum number of turns becomes a compromise. I have no choice but to follow.

Therefore, as a way to solve these problems,
As shown in Figure 4, the control winding is divided into a winding with a large number of turns and a winding with a small number of turns.The winding with a large number of turns is directly driven by a control amplifier, and the winding with a small number of turns is driven directly by a control amplifier. An effective method is to drive via a capacitor. In Fig. 4, 32 to 34 are control windings with a small number of turns, 35 to 37 are control windings with a large number of turns, and 3
8 is a frequency division capacitor.

By performing control as shown in Figure 4,
For signals with high frequency components and sharp rises, the control windings 32 to 34 with a small number of turns are driven mainly through the capacitor 34, so the response is quick and the slew rate can be increased, and signals with low frequency components For signals, since the control windings 35 to 37 having a large number of turns are driven, only a small amount of control power is required, so that both response speed and power efficiency can be achieved.

Although the embodiment of FIG. 1 shows an example of an amplifier with one set of input and output, that is, one channel, the square wave voltage generator can be shared even in the case of multiple channels. In this case, compared to the case where a square wave voltage generator is provided in each channel, there is an advantage that beats due to the frequency difference between the square wave voltage generators do not occur.

Next, the structure of each transformer will be explained.
The transformers 2 and 3 and the transformer 4 can have basically the same structure except for the presence or absence of the secondary winding and the transformer winding. For example, an example of the structure of the transformer 2 is shown in FIG.

In the example shown in Figure 5, an E/I type core is used, the primary winding and secondary winding are wound around the central magnetic path, and the control winding and bias winding are wound around the magnetic paths on both sides. Each piece is divided into two parts and rolled up.

In Fig. 5, 39 is the central magnetic path, 40, 4
1 is a magnetic path on both sides, 42 and 43 are control windings, 44,
45 is a bias winding.

Control winding 4 wound around magnetic paths 40 and 41 on both sides
The magnetic flux produced by the bias windings 44 and 45 circulates only through the magnetic paths 40 and 41 on both sides, changing the magnetic resistance value of the loops. Also, the magnetic flux created by the primary winding 5 wound around the central magnetic path flows from the central magnetic path 39 to the magnetic paths 40 and 4 on both sides.
It splits into 1 and returns to the central magnetic path. As a result, the inductance of the primary winding 5 is changed by the magnetic resistance of the magnetic paths on both sides. In this way, the primary winding 5 is
The inductance can be changed.

With this configuration, the voltage induced in the control windings 42 and 43 or the bias windings 44 and 45 by the square wave voltage applied to the primary winding 5 is as follows.
There is an advantage that the polarities are opposite to each other between 42 and 43 or between 44 and 45, and because they are added, they are canceled and do not appear outside the winding.

In addition, in the case of a structure in which all the windings are wound on the same magnetic path, the magnetic flux due to the square wave is added to the magnetic flux due to the control winding and bias winding when the polarity is positive or negative, and when the polarity is opposite As a result, the inductance value of the primary winding for the positive polarity and negative polarity of the square wave is different, and as a result, the voltage of the secondary winding during the period when diode 19 is conducting and the period when diode 20 is conducting There is a problem that the ripple of the square wave component of the output increases because the voltage of the secondary winding of the Since it is added in one magnetic path and subtracted in the other magnetic path,
It has the great advantage of being averaged in both directions, making the inductance equal in both directions, resulting in very low output ripple.

In addition, the core of the transformer is E and I in the example shown in Figure 5.
Although a mold is used, it is exactly the same even if E/E types are used.

As explained above, according to the magnetic amplifier of the present invention, efficiency can be significantly improved compared to conventional class A and class B amplifiers, and since the amplifier itself has an isolation function, a power transformer can be omitted. ,
The cost of the radiator and power transformer can be reduced, and in the case of battery drive, an arbitrary output can be obtained by changing the primary and secondary winding ratios.

[Brief explanation of the drawing]

Figure 1 is a circuit diagram showing one embodiment of the present invention, Figure 2 is a circuit diagram showing an embodiment of the present invention.
3 is a diagram for explaining the operation of the present invention. FIG. 4 is a circuit diagram showing another embodiment of the control section in the present invention. The figure is a block diagram of a transformer that can be used in the present invention. 1...Square wave voltage generator, 2...First transformer, 3...Second transformer, 4...Third transformer, 5-7...Primary winding, 8, 9...Secondary Winding, 10-12... Control winding, 13, 14... Bias winding, 15, 16... Bias supply circuit,
17... Control amplifier, 19, 20... First rectifier circuit, 21, 22... Second rectifier circuit, 25,
26...Means for combining the outputs of the first and second rectifier circuits and supplying them to the load, 27...Ripple removal capacitor, 28, 29...Negative feedback circuit, 32-
34... Control winding with a small number of turns, 35-37... Control winding with a large number of turns, 38... Frequency division capacitor, 39... Central magnetic path, 40, 41... Magnetic paths on both sides, 42 , 43... control winding, 44, 45...
...bias winding.

Claims (1)

[Claims] 1 connected to a series connection circuit of first and second transformers each having a primary winding, a secondary winding, a control winding, and a bias winding, and the primary windings of the first and second transformers. a square wave voltage generator; first and second rectifier circuits respectively connected to the secondary windings of the first and second transformers; a control amplifier that drives the line, a bias current supply circuit that supplies bias current to the bias windings of the first and second transformers, and the outputs of the first and second rectifier circuits that are combined and supplied to the load. The polarity of the control winding and bias winding of one of the first and second transformers is in the direction of adding magnetic flux, and the polarity of the other is in the direction of subtracting magnetic flux, and A magnetic amplifier, wherein one of the first and second rectifier circuits is polarized so that it generates a positive voltage, and the other one generates a negative voltage.