JP2502967B2 - Switching regulator - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a switching regulator, and more particularly to a switching regulator suitable for a push-pull type, a half-bridge type and the like.

BACKGROUND OF THE INVENTION Switching regulators are largely classified into flyback type, forward type, and push-pull type based on the magnetization characteristics of a transformer.

Among them, the push-pull type is widely used for models with a medium capacity or more because of the high utilization rate of the transformer core and the small current required for the switch element. In addition, demand for power sources for computer-related equipment is increasing year by year.

However, in the push-pull type, as is well known, since the main transformer is demagnetized, there are not a few accidents such as switch element destruction. Therefore, as described in JP-A-59-6773,
Bias prevention measures are considered to be important issues. However, this technique has a problem that the circuit loss becomes large because the main circuit requires a resistor for current detection.

Further, the time constant of the current detection circuit needs to be made sufficiently larger than the operating frequency of the main switch element, the response speed to bias magnetism deteriorates, and there is a risk that bias magnetism cannot be suppressed when there is a sudden change in input. . In general, the switching element is controlled so as to detect the exciting current of the main transformer during the ON period of the switching element and correct the irregularity of the Vt product.
For example, as shown in JP-A-52-74828, the circuit becomes complicated. Further, the maximum output is the most severe condition for the magnetic bias, but at this time, since the exciting current is extremely small with respect to the load current, sufficient detection sensitivity cannot be obtained.

Further, since the switch element is driven by the feedback signal on the output side so that the output is stabilized, the accuracy of output stabilization is impaired when the bias detection signal of the main transformer is added.

For this reason, “Asymmetry of Switching Impulse-Wise-Moderated Power Surprise” (Electron; 167 p33-34, GBR,
1979) (Asymmetry of Sw-itching in Pulse-Width-
Modulated Power Supplies (Electron; No.167, Page33〜
34, GBR, 1979] and a special issue on power generation-SW regulators (transistor technology, February 1979, pages 208 to 209), which often employ half-bridge or full-bridge type.

These have a compensating capacitor in series with the input winding of the main transformer. For this reason, it is necessary to use an expensive and large-sized capacitor having a capacity and a high breakdown voltage that can cover the converted power.

Further, for the purpose of controlling the output to be constant, Japanese Patent Laid-Open No. 57-1117
It is known that a saturable reactor is provided on the output side of a main transformer as shown in Japanese Patent No. 13 publication. In this case, a freewheeling diode is provided to regenerate the energy stored in the chiyoke coil during the off period of the switch element so that this current does not flow to the saturable reactor. This is because it does not adversely affect the reset of the saturable reactor. Also shown is the use of a saturable reactor with a main winding on one tripod core and a reset winding biased by an external DC power supply. In this circuit configuration, the saturable reactor is biased on one side only, and the main transformer is always magnetized in one direction via the main transformer winding. However, since the magnetic flux density of the main converter iron core changes from the first quadrant to the third quadrant of the BH curve in both the positive and negative directions, it operates when magnetized in only one direction by the saturable reactor. Apart from the control function, the magnetic bias is always increased and reaches saturation. For this reason, the present technology further requires a means for preventing the magnetic bias of the main transformer.

[Object of the Invention]

An object of the present invention is to provide a switching regulator that can prevent biasing and saturation of the main transformer with a simple circuit configuration without sacrificing output accuracy.

[Outline of Invention]

The present invention focuses on the fact that the exciting current of the main transformer during the OFF period of the switch element is released to its output side, and a bias magnetizing correction circuit is provided in a part of the circuit configuration on the output side of the main transformer. It is in. This bias magnetizing correction circuit generates a voltage only during a time corresponding to the magnitude of the exciting current during the switch element off period,
This voltage is configured to be applied to the winding on the output side of the main transformer. At this time, since the iron core of the main transformer can be magnetized to have a polarity opposite to that of the previous ON period, it is possible to autonomously correct the magnetic bias without using an external circuit.

Example of Invention

An embodiment of the present invention will be functionally described with reference to FIG.
Reference numeral 1 is an input DC power supply, 2 is a main transformer, 3 is a switching unit, 4 is a drive circuit, and 8 is a control circuit. The eccentricity correction circuit 5 and the rectification circuit 6 and the output smoothing circuit 7 are also provided on the output side of the main transformer.

The bias correction circuit 5 generates a voltage for the time corresponding to the unbalanced portion of the exciting current and applies it to the main transformer 2 while all the switching elements of the switching unit 3 are off. This voltage is generated so that the main transformer 2 has a polarity opposite to the polarity magnetized during the ON period of the switch element.

Next, a concrete embodiment of the present invention will be described with reference to FIG. Q ₁ and Q ₂ are switching elements such as a pair of transistors and are the main components of the switching unit 3. Transformer 2 is
Intermediate tap type primary and secondary windings are provided, and a bias magnetizing correction circuit 5, a rectifying circuit 6 and an output smoothing circuit 7 are connected to the secondary winding. The overall configuration is a push-pull type switching regulator and can supply a predetermined electric power during the ON period of the switching elements Q ₁ and Q ₂ .

The bias correction circuit 5 can be configured by a saturable reactor M _L. The windings W ₅₁ and W ₅₂ wound around the same iron core are connected in series with the diodes D ₁ and D ₂ of the rectifier circuit 6, respectively, so that the current path of the secondary winding of the main transformer 2 flows. Becomes

The operation of the embodiment of this configuration will be described by dividing it into the following four modes. However, the drive circuit 4 for the switch elements Q ₁ and Q ₂ and the control circuit 8 shown in FIG. 1 are conventional techniques, and a detailed description thereof will be omitted.

(1) Mode 1 is the period during which Q _{1 is} on and Q _{2 is} off.

During this period, from the secondary winding W ₂₁ of the main transformer 2 to W ₅₁ , D ₁
And output terminal 9 via the chioke coil L _F and the capacitor C _F.
Power is supplied to the load from a and 9b. At the same time, magnetic energy is accumulated in the chioke coil L _F.

Explaining the operation of the saturable reactor M _L during this period,
Since the voltage of the secondary winding W ₂₂ of the main transformer ₂ is blocked by the diode D ₂ , no voltage is applied to the winding W ₅₂ . One W
Immediately after the switch element Q ₁ is turned on, the voltage ₅₁ induces a voltage having a black circle in the drawing as a positive polarity, but immediately reaches the magnetic saturation region and exhibits a low impedance, enabling the power supply as described above. At this time, the magnetization direction of the saturable reactor M _L is set to the positive direction.

(2) Mode 1 is the period of Q ₁ ; off, Q ₂ ; off.

First, the operation when the saturable reactor 5 is not added, that is, when the terminal of the black circle shown in W ₂₁ is directly connected to the diode D ₁ and the terminal opposite to the black circle of W ₂₂ is directly connected to the diode D _2. I will describe.

During this period, Q ₁ turns off, so that the exciting current that was flowing from DC power supply 1 through primary winding W ₁₁ and Q ₁ of the main transformer in mode 1 flows through the same path in mode 2. I can't. Therefore, consider the path in which this exciting current flows in mode 2. The exciting current flowing in the W ₁₁ is, W ₁₁
Any winding can flow if it is magnetically coupled to the winding. However, the polarity of the exciting current must flow from the terminal with the opposite polarity to the black circle in the figure, as in the case of flowing through W ₁₁ .

First, the exciting current flowing in the winding W ₁₁ is 2
Consider the case where the current flows through the next winding W ₁₂ . At this time, the exciting current is W
₁₂ black circles and terminals of opposite polarity, DC power supply 1, anti-parallel diode of Q ₂ and W (not shown) actually provided
It will flow through the path of the ₁₂ black terminals. Therefore, the voltage of the input DC power supply 1 is caused in the winding W ₁₂ by setting the polarity opposite to that of the black circle in the drawing to be positive.

On the other hand, consider the case where the exciting current flows through the secondary winding side of the main transformer.

In the mode 2, the exciting current (the converted value on the secondary side is i _ex ) flows out from the winding W _22, and flows through the diodes D ₂ and D ₁ to the path of the black circle of the winding W ₂₁ . At this time,
In addition to the exciting current, the load current is flowing through the diode D ₁ , and the exciting current is sufficiently smaller than the load current, so that the diode D ₁ is not turned off. Since the impedance of this path is smaller than the impedance of the exciting current path on the primary winding side, in mode 2, the exciting current flows through the secondary winding side.

If the load current flowing in the chain yoke coil L _F in mode 1 is 2 I ₀ , this current flows into the center tap of the secondary winding via the capacitor C _F and the load, and then the winding W ₂₁
And flow into W ₂₂ equally. That is, the current I ₀ flows through the windings W ₂₁ and W ₂₂ , respectively. Therefore, considering the exciting current i _ex , the current flowing through the diode D _{1 in} mode 2 is i ₁ = I ₀ −i _ex (1) Further, the current flowing through the diode D ₂ is i ₂ = I ₀ + i _ex ... (2)

Next, the operation when the saturable reactor 5 is added will be described.

Saturable reactor magnetic path length L _m , coercive force H _c , winding W
_Assuming that the turns of ₅₁ and W ₅₂ are n _M , respectively, even if an attempt is made to magnetize the saturable reactor with a current larger than the current i _M shown in the following equation (3), a current greater than the current i _M cannot flow, and the saturable reactor is Presents a large impedance. That is, the saturable reactor can be equivalently regarded as a current source of the current i _M.
i _M = L _m · H _c / (2n _M ) ... (3) Therefore, when a saturable reactor is provided, in mode 2 only the current i _M flows in the secondary winding in addition to the load current, and the exciting current The difference current (i _ex −i _M ) between i _ex and the current i _M flows out from the primary winding W ₁₂ of the main transformer. Accordingly, the voltage of the input DC power source 1 is positively induce black circle opposite polarity side of the windings W _12. This voltage is the opposite of the polarity of the voltage applied to W ₁₁ in mode 1 and will magnetize the transformer in the opposite polarity to mode 1.

Also, the current (i _ex −i _M ) flows into the input DC power supply 1,
The magnetic energy stored in the main transformer in Mode 1 is regenerated to the input DC power supply and gradually decreases to i _M. And when the exciting current of the main transformer decreases to i _M ,
After that, the saturable reactor becomes low impedance again. Therefore, the secondary side of the main transformer is short-circuited, and the winding edge W _{51 of the} saturable reactor has (I ₀ −i _M ), winding W
The current of (I ₀ + i _M ) continues to flow in ₅₂ and reaches a steady state. Further, the exciting current flowing in the primary side becomes 0, and the voltage of the input DC power source induced in the winding W ₁₂ also becomes 0.

From the above operation, the main transformer magnetized in mode 1
It can be seen that in mode 2, the secondary side converted value of the exciting current is automatically reset until it becomes equal to the holding force of the saturable reactor.

(3) Mode 3 is the period during which Q _{1 is} off and Q ₂ is on.

During this period, a voltage having a positive polarity on the side opposite to the black circle in the figure is induced in the secondary winding W ₂₂ of the main transformer 2 and the main transformer 2
In contrast to mode 1, the magnetic flux density changes in the negative direction. Similarly, the saturable reactor M _L also changes in the negative direction and becomes saturated,
Electric power is supplied via W ₅₂ and D ₂ .

(4) Mode 4 is the period of Q ₁ ; off, Q ₂ ; off.

During this period, the differential current (i _ex −i _M ) is the primary winding W of the main transformer.
Q ₂ which flows out from ₁₁ and is actually provided though not shown
Flow through the anti-parallel diode. As a result, the input DC power supply 1 induces the polarity of the black circle of the winding W ₁₁ into a positive voltage. This voltage is the opposite of the polarity of the voltage applied to W ₁₂ in mode 3 and will magnetize the transformer in the opposite polarity to mode 3.

Also, the current (i _ex −i _M ) flows into the input DC power supply 1,
The magnetic energy stored in the main transformer in mode 3 is regenerated to the input DC power supply and gradually decreases to i _M. And when the exciting current of the main transformer decreases to i _M ,
After that, the saturable reactor becomes low impedance again. Therefore, the secondary side of the main transformer is short-circuited, and the winding W ₅₁ of the saturable reactor has (I ₀ + i _M ), winding W
The current of (I ₀ −i _M ) continues to flow in ₅₂ and reaches steady state. Further, the exciting current flowing in the primary side becomes 0, and the voltage of the input DC power source induced in the winding W ₁₁ also becomes 0.

From the above operation, the main transformer magnetized in mode 3
It can be seen that in mode 4, the secondary side converted value of the exciting current is automatically reset until it becomes equal to the holding force of the saturable reactor.

As described above, in the present embodiment, the bias magnetism can be corrected in response to only the exciting current, regardless of the load current.

In addition, since the magnetic bias of the main transformer occurs due to a slight imbalance of the Vt product, the Vt product of the saturable reactor M _L can be designed to be small in this embodiment.

Therefore, as described in Mode 1 and Mode 3 above, the saturable reactor M _L immediately after the main switch element is turned on.
The non-conduction period of can be made extremely short.

The operation of the saturable reactor M _L of FIG. 2 will be described in more detail here.

First, mode 1 is the ON period of Q _1, shown in W ₂₁ ●
A voltage having a positive polarity is induced and power is supplied to a load (not shown) via the diode D ₁ . At this time, in the iron core of the main transformer 2, the magnetic flux density changes to point A as indicated by an upward arrow in the B-H curve of FIG.

Next in mode 2, is off periods of the two transistors, a current 2I was flowing Chiyokukoiru L _{F 0} is the winding W ₂₁
Since flow is divided into a W _22, cancel each other and the magnetic flux density from each other, the winding of the main transformer 2 no voltage. That is, when viewed on the magnetization curve, it stops at point A.

Mode 3 is the ON period of the switch element Q ₂ , in which a voltage is applied to W ₁₂ in the opposite polarity to the mark ● in the figure, and power is supplied from W ₂₂ via diode D ₂ . At this time, the magnetization of the main transformer 2 advances to the point B as shown by the downward arrow in FIG.

Next, in mode 4, the two yokes Q ₁ and Q ₂ are turned off again, and in the same manner as in the case of mode 2 above, the yoke coil is used.
The current of L _F splits into W ₂₁ and W ₂₂ and stops at point B on the magnetization curve to complete one cycle.

When the Vt products of the primary windings W ₁₁ and W ₁₂ operate in an ideally balanced manner, the main transformer is magnetized to ± B _m and no bias is generated.

On the other hand, the saturable reactor M _L normally operates as just a mechanical switch. That is, in mode 1, when a positive voltage is applied to the mark ● of winding W ₅₁ , it is magnetically saturated and turned on, so that the current of W ₁₁ flows to the load side. .

Next, in the mode 2, the switch element Q ₁ is turned off, and at this time, the exciting inductance of the main transformer 2 emits exciting energy having a positive polarity opposite to the mark ● in the figure. As can be seen from FIG. 2 described above, this excitation energy cannot flow to the primary winding side, and eventually flows to be superimposed on the current of the chioke coil L _F. Chiyokukoiru L current i ₁ flowing through the W ₅₁ from the current 2I ₀ does not change the _F is an expression (2), also W ₅₂
Current i ₂ of the above equation (1).

Therefore, the saturable reactor M _L is relatively i ₂ −i ₁ = 2i
_Since it is magnetized with the current of _{ex to} the polarity opposite to that shown by ●, if the number of turns of windings W ₅₁ and W ₅₂ of saturable reactor M _L is the same, then n. Is magnetized in the opposite direction, and as shown in FIG. 3 (b), it changes from point Q to point P. At this time, if the core magnetic path length l of the saturable reactor M _L and the windings W ₅₁ and W ₅₂ are selected so that the magnetic flux density does not change, no voltage is induced in the winding of the saturable reactor M _L. .

Next, the process of mode 3 and mode 4 is symmetrically magnetized in the opposite direction to the above, and completes one cycle through Q and R points.

Here, the operation of the present invention will be described in the case where the bias magnetism occurs in the mode 1. In this case, in the magnetization curve of FIG. 3 (a), the operating point A increases and becomes larger than B _m and H _m , so the exciting current emitted in mode 2 naturally becomes large, and the saturable reactor M _L becomes Reverse excitation is performed deeper than in the normal state and reaches the H ₀ point in FIG. 3 (b). At this time, a voltage of V = n · S · ΔB / t ((5)) is induced in the winding of saturable reactor M _{L in} the opposite polarity to the mark ● in the figure, so this voltage is applied to diodes D ₁ and D ₂
And applied to W ₂₁ and W ₂₂ of the main transformer 2 via the output smoothing circuit 7, and the magnetic flux density decreases toward point A in FIG.

Even if the demagnetization occurs in the mode 3, the demagnetization of the main transformer 2 can be suppressed by the voltage induced in the saturable reactor M _L by the same principle.

Since the magnetic bias of the main transformer 2 occurs due to a slight imbalance of the Vt product, it is possible to obtain sufficient detection sensitivity by using a saturable reactor having a large magnetic permeability, as shown in Fig. 3 (b). Can be corrected.

As another embodiment of the present invention, the M _L iron core can be implemented by a normal reactor having no square characteristic. Although various shapes of the iron core can be used, a small iron core is sufficient as described above.

The M _L using this type of iron core has a voltage even during normal operation of the main transformer 2, but the objective can be achieved by designing the operating point at the time of eccentricity to a large change on the BH curve. To be

In this embodiment, the iron core can be obtained at a lower price than the saturable reactor, and the coercive force is small, so that the hysteresis loss is small. Therefore, there is an advantage that it can be applied to high frequency driving.

FIG. 4 shows another embodiment of the present invention. The present invention can be applied to an output in which the secondary winding as shown in the figure is not of the intermediate tap type.

In FIG. 4, W ₂₀₁ and W ₂₀₂ are independent secondary windings and have the same number of turns. In this embodiment, the common electrode of the diode of the rectifier circuit 6 can be connected to the output terminal 9b and provided.

The principle of operation is exactly the same as that described in FIG. 2 above, and the current of the chioke coil L _F flows separately in i ₁ and i ₂ in mode 2 (or mode 4), so the demagnetization correction circuit 5
The voltage that corrects the bias magnetism is obtained by the difference between the exciting currents.

It is known that the diodes D ₁ and D ₂ usually have a capacitance C _S with respect to virtual ground because the common electrode side is usually attached to the cooling fin. In this embodiment, since the potential is fixed at one end of the DC output, there is an advantage that noise due to charging / discharging of C _S does not occur.

Next, another embodiment of the two-terminal output main transformer as shown in FIG. 5 will be described.

Since the voltage of the secondary winding W ₂₀ is alternately inverted by turning on and off of Q ₁ and Q ₂ , four rectifying diodes are required as shown in FIG.

In this embodiment, the first on-period or mode _{1 Q 1, D 11, L} F, C F, power is supplied through the winding W _53, D ₂₂ of the reactor M _L. At this time, M _L is magnetized by a voltage having a black circle in the figure as a positive polarity and saturated in the positive direction.

Next, in the mode 2 in which both switches Q ₁ and Q ₂ are turned off, the current of the chuck yoke coil L _F can flow through the path of D ₂₁ , D ₂₂ or D ₁₁ , D ₁₂ , but on the D ₂₁ and D ₂₂ side. Since the resistances of W ₅₃ and W ₅₄ exist in series, they eventually flow through D ₁₁ and D ₁₂ .

Further, in the mode 2, the exciting current i _ex is emitted from the winding W ₂₀ of the main transformer 2 with the side opposite to the black circle shown in the drawing having a positive polarity.

This current can flow in path 1 because D ₁₁ is activated. Since the saturable reactor M _L in the excitation current i _ex it is magnetized in the mode 1 and the reverse direction, appeared voltage the side opposite to the illustrated black circles and the positive polarity is applied to the windings W ₂₀ of the main transformer 2 . In mode 3 and mode 4, the same operation is performed symmetrically, and it is possible to prevent the main transformer 2 from being demagnetized.

In the present embodiment, since the power conversion can be obtained with only one output winding of the main transformer 2, there is an advantage that the winding configuration of the main transformer 2 can be simplified.

Although all of the above embodiments have been described with respect to the push-pull type, the full-bridge type shown in FIG. 6 can also be applied. The output side of the main transformer 2 is shown in FIG. In this embodiment, since the bias magnetism can be corrected on the output side, an expensive compensating capacitor C ₀ is not necessary and 1
It is possible to take full advantage of the full-pledge type, which does not require an intermediate tap of the next winding.

FIG. 7 shows a half-bridge type embodiment. In this case as well, C ₀ is unnecessary, and even if there is a capacitance imbalance between the split capacitors C ₁ and C ₂ or there is a difference in leakage current, bias magnetism can be corrected on the output side, so the original advantage is fully utilized. .

In addition, switching regulators often include a so-called multi-output type in which a plurality of independent outputs are obtained from one main transformer. These may be provided with a system for providing a bias magnetic correction circuit only for the main output with large power, or for each output. In consideration of the impedance of each output, the bias magnetizing correction circuit 5 may be provided at the output where the exciting current flows most during the OFF period of the switch element.

In the above-described embodiments, the output smoothing circuit 7 is set to L,
Are composed of C circuits, for example, aims a constant current output, it may be constituted only Chiyokukoiru L _F. In this case, since the exciting current in the mode 2 or the mode 4 flows through the load, the same operation as that of the above embodiment can be obtained.

〔The invention's effect〕

According to the present invention, it is possible to obtain the voltage for correcting the bias magnetism by utilizing the exciting current discharged to the output side of the main transformer, so that it is possible to prevent the bias magnetism and the saturation phenomenon of the main transformer with a simple circuit configuration.

[Brief description of drawings]

FIG. 1 is a block diagram showing a circuit configuration of the present invention, FIG. 2 is an embodiment, FIG. 3 is a B-H curve of the main transformer 2 and the saturable reactor M _L , and FIGS. It is a circuit diagram which shows the other Example of invention. DESCRIPTION OF SYMBOLS 1 ... Input DC power supply, 2 ... Main transformer, 3 ... Switching part, 5 ... Demagnetization correction circuit, 6 ... Rectifier circuit, 7 ... Output smoothing circuit.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Kenichi Onda 4026 Kujimachi, Hitachi, Ltd. Hitachi, Ltd., Hitachi Research Laboratory (72) Inventor Kohei Yabino 4026 Kujicho, Hitachi, Ltd. Hitachi, Ltd. Hitachi Research In-house (56) References JP-A-51-62315 (JP, A) JP-A-57-55771 (JP, A) Actual opening 60-24190 (JP, U) Actual opening Sho-58-31789 (JP, U) US Patent 4217632 (US, A)

Claims (1)

(57) [Claims] 1. An input DC power supply, a conversion circuit for AC-exciting a primary winding of a main transformer by ON / OFF control having a period in which the input DC power supply is simultaneously turned off by at least a pair of switch elements, and the main transformer. A full-wave rectification circuit for performing full-wave rectification by a diode and a smoothing circuit including a choke coil and a capacitor for smoothing the output of the full-wave rectification circuit are provided on the secondary winding side of the Which is a push-pull type switching regulator,
A biased magnetism that applies a voltage corresponding to the magnitude of the exciting current of the primary winding of the main converter between the secondary winding of the main converter and the smoothing circuit to the secondary winding of the main converter. In the switching regulator to which a correction circuit is connected, the bias correction circuit is composed of two saturable reactors that are wound around the same iron core and are magnetically coupled to each other. The secondary winding of the main converter formed for each rectification mode of the full-wave rectification circuit, the diode of the full-wave rectification circuit, and the choke coil are connected in series in a closed loop. The two saturable reactors are magnetically coupled so as to cancel the magnetic flux formed by each other with respect to the current flowing from the choke coil and strengthen the magnetic flux with respect to the exciting current of the main converter. Switching regulator to be.