JP6565770B2 - Push-pull DC / DC converter - Google Patents

Description

  The present invention relates to a push-pull type isolated DC / DC converter that alternately operates a plurality of primary side switching elements and converts electric power through a transformer.

2. Description of the Related Art Conventionally, there is known a power conversion device that converts power by switching on and off a plurality of switching elements. In this type of power conversion device, when the switching element is turned on and energization is started before the drain-source voltage of the switching element decreases, a turn-on loss due to the product of the current and voltage occurs.
For example, in the push-pull DC / DC converter disclosed in Patent Document 1, the voltage of the main switch is lowered by operating the auxiliary switch and charging / discharging the external capacitor using the primary side leakage inductance. The structure that turns on afterwards is adopted.

International Publication WO2008 / 020629

In the technique of Patent Document 1, an auxiliary switch is required to charge and discharge the charge of the external capacitor, and there is a problem that the number of parts increases and the physique increases.
The present invention was created in view of the above points, and an object thereof is to provide a push-pull DC / DC converter that reduces the turn-on loss of the primary side switching element without using an auxiliary switch. .

  The present invention relates to a push-pull DC / DC converter that is connected between a power source (Bt) and a load (Ld) and converts DC power. This push-pull type DC / DC converter includes a plurality of primary coils (31, 32) and a plurality of secondary coils (33, 34) constituting an exciting inductor of the transformer (20), and a plurality of primary side switching elements (SW1). , SW2), a plurality of secondary rectifier elements (SW3, SW4, DI3, DI4), a smoothing inductor (7) connected between the secondary rectifier elements and the load, and a switching time calculator (13 ) And a switching operation unit (18).

The plurality of primary side switching elements are connected between the plurality of primary coils and the power source, and the return diodes are connected in parallel, and operate alternately.
The plurality of secondary-side rectifying elements are connected to the plurality of secondary coils and can rectify the current flowing through the secondary coils. For example, the secondary side rectifying element is a diode or a switching element.
In the configuration where the secondary-side rectifying element is a switching element, it is assumed that the secondary-side switching element in the on state is turned off before shifting from the normal state described later to the reverse state.

The switching time calculation unit includes a theoretical on-time (Ton) that is “a time during which any primary-side switching element is turned on and the load current (i _L ) flowing through the load and the smoothing inductor is theoretically increased”, and “all The theoretical off time (Toff), which is the time during which the primary side switching element turns off and the load current theoretically decreases, is calculated.
Here, “calculating the theoretical off time” means that the theoretical off time is not simply obtained by subtracting the theoretical on time from the switching period set as a fixed value, but based on some technical idea. It means calculating the time when the current theoretically decreases.
The switching operation unit operates the primary side switching element based on the theoretical on time and the theoretical off time calculated by the switching time calculation unit, and further operates the secondary side switching element when the secondary side rectifying element is a switching element. .

Hereinafter, in the period in which the load current when all the primary side switching elements are off flows back through the secondary side rectifier element, the load current that flows back per secondary side rectifier element is referred to as element load current (i _L / 2). To do. Further, regarding the excitation current flowing through the excitation inductor of the transformer, the excitation current converted to the current flowing per secondary side rectifying element is defined as the converted excitation current (i _Lm / 2N).
Also, the state where the absolute value of the element load current is equal to or greater than the absolute value of the converted excitation current is defined as `` normal state '', and the state where the absolute value of the element load current is lower than the absolute value of the converted excitation current Is defined.

After switching from the normal state to the reverse rotation state, the switching operation unit turns on the primary side switching elements in the next turn-on state during the reverse rotation state.
When the exciting current flowing on the secondary side starts to commutate to the primary side, the charge of the capacitance between the drain and source of the primary side switching element is discharged, and the drain-source voltage drops. Moreover, it is in a reverse state at the timing when the theoretical off time has elapsed.
Therefore, the switching operation unit of the present invention reduces the voltage immediately after the start of energization by turning on the primary side switching element during the reverse rotation state based on the theoretical on time and the theoretical off time calculated by the switching time calculation unit. Can be suppressed. That is, in the present invention, by using the excitation current, the turn-on loss can be reduced with a simple configuration without adding an auxiliary switch which is a separate part as in the prior art.

  By the way, as a configuration for determining the reverse rotation state in addition to the determination from the theoretical off time, for example, it may be determined by detecting a current flowing in the primary side switching element or the secondary side rectifying element or a change in both-end voltage. However, in order to detect these currents and voltages, it is necessary to add a current sensor and a voltage sensor exclusively for detecting the reverse state. It is also necessary to ensure sensor accuracy.

In contrast, in the present onset bright, the switching time calculation unit calculates the theoretical off-time by the following equation.
Toff = [{(i _L −i _Lm ) / N} / 2v _out ] × L
The meaning of each symbol is as follows.
Toff: Theoretical off time
(I _L / 2): Peak value of element load current
(I _Lm / 2N): Conversion excitation current when all primary side switching elements are off in a configuration in which the winding ratio between the primary side and the secondary side of the transformer is 1: N
v _out : Output voltage output to the load when all primary side switching elements are off
L: Inductance of the smoothing inductor These pieces of information are originally necessary information for controlling the basic operation of the push-pull type DC / DC converter. Therefore, by determining the reverse rotation state based on the theoretical off time calculated based on these pieces of information, it is not necessary to add a sensor dedicated to detection, and it is not necessary to consider the influence of sensor accuracy.

Preferably, the switching operation unit turns on the primary side switching elements in the next turn-on order after the drain-source voltage of the primary side switching elements drops and reaches zero voltage. Thereby, “zero voltage switching” can be realized, and the turn-on loss can be particularly reduced.
For this purpose, the switching time calculation unit further calculates an adjustment off time (Toff ^* ) obtained by adding a predetermined extension time (Tadd) to the theoretical off time. It is preferable to turn on the primary side switching elements in the turn-on order. By appropriately setting the extension time, zero voltage switching can be reliably realized.

The block diagram of the push pull type DC / DC converter whose secondary side rectification element used for each embodiment is a diode. The block diagram of the push pull type DC / DC converter whose secondary side rectifier used for each embodiment is a switching element. The time chart which shows operation | movement of a push pull type DC / DC converter. The secondary side rectifier is a diode or switching element and the synchronous rectification control is stopped before the reverse rotation start time. Time chart shown. FIG. 5 is a diagram of current paths in modes (a) and (b) in the configuration of FIG. FIG. 5 is a diagram of current paths in modes (d), (e), and (f) in the configuration of FIG. 4. The time chart which shows the change of the electric current and voltage of each part at the time of a primary side switching element ON transition (load current minimum area | region) by the structure which a secondary side rectifier is a switching element and continues synchronous rectification control after the reverse rotation start time. FIG. 8 is a diagram of current paths in modes (a), (b), and (c) in the configuration of FIG. 7. [A]: The figure which added the voltage symbol to the mode (d) of FIG. [B]: Circuit model diagram. (A) The block diagram of the switching time calculation part of 1st Embodiment. (B) The figure which shows the relationship between the change of load current, a theoretical on time, and a theoretical off hour meter. The block diagram explaining Duty control of a well-known technique. The block diagram of the switching time calculation part of 2nd Embodiment. The time chart which shows the change of the load current when not setting extension time. The time chart which shows the change of load current at the time of setting extension time. The figure which shows the effect of zero voltage switching. The figure which shows the effect of zero voltage switching.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. In a plurality of embodiments, substantially the same configuration is denoted by the same reference numeral and description thereof is omitted. Further, the first and second embodiments described below are collectively referred to as “this embodiment”.
First, two configuration diagrams of the push-pull type DC / DC converter are shown in FIG. 1 and FIG. These are different only in the configuration of the secondary side rectifying element, and the other configurations are common. The push-pull type DC / DC converter 101 shown in FIG. 1 uses diodes DI3 and DI4 as secondary side rectifying elements. On the other hand, in the push-pull DC / DC converter 102 shown in FIG. 2, switching elements SW3 and SW4 are used as secondary side rectifying elements.

  However, the feature of each embodiment is not the configuration itself of the secondary side rectifying element but a configuration in which the “switching time calculation unit” particularly calculates the theoretical off time Toff. The configuration of the diode as the secondary side rectifying element and the configuration of the switching element can be applied to any of the embodiments. Thus, first, the configuration of the two push-pull type DC / DC converters 101 and 102 will be described, and then the description will proceed to each embodiment.

First, a push-pull type DC / DC converter 101 having secondary-side rectifying elements of diodes DI3 and DI4 will be described with reference to FIG. The push-pull type DC / DC converter 101 is connected between the power source Bt and the load Ld, and converts DC power.
The push-pull type DC / DC converter 101 includes two sets of primary coils 31, 32 and secondary coils 33, 34, two primary side switching elements SW1, SW2, and two secondary side diodes that constitute an exciting inductor of the transformer 20. DI3, DI4, smoothing inductor 7 and the like are provided. Further, the push-pull type DC / DC converter 101 includes a switching operation unit (“SW operation unit” in the figure) 18 for operating the primary side switching elements SW1 and SW2, and a switching time calculation unit (“SW time in the figure). Calculation unit ") 13.

The transformer 20 is configured such that the turns ratio between the primary side and the secondary side is “1: N”, and the exciting current I _Lm flows through the exciting inductor. Exciting current I _Lm is equal to the sum of the current _{I t1, I t2, N ×} I t3, N × I t4 flowing through each coil 31, 32, 33, 34. In the present embodiment, description will be made mainly assuming the case where the turns ratio is 1: 1.
One ends of the primary coils 31 and 32 are connected by a primary side center tap CT1, and the center tap CT1 is connected to the positive electrode of the power source Bt. The other ends of the primary coils 31 and 32 are connected to the negative electrode of the power source Bt via primary switching elements SW1 and SW2, respectively. The input voltage input from the power supply Bt is represented as “v _in ”.

  One ends of the secondary coils 33 and 34 are connected by a secondary center tap CT2, and the center tap CT2 is connected to one terminal Pp of the load Ld. The other ends of the secondary coils 33 and 34 are connected to the cathodes of the diodes DI3 and DI4, respectively. Diodes DI3 and DI4 as secondary rectifiers rectify the current flowing through the secondary coils 33 and 34.

Between the anodes of the diodes DI3 and DI4 and the other terminal Pn of the load Ld, a smoothing inductor 7 that stores magnetic energy by energization is connected. The inductance of the smoothing inductor 7 is expressed as “L”.
A current flowing through the load Ld and the smoothing inductor 7 is referred to as a load current i _L. The load current i _L has a positive direction from the load Ld to the diodes DI3 and DI4 via the smoothing inductor 7. A secondary-side smoothing capacitor 8 is connected in parallel with the load Ld.
The output current output to the load Ld is represented as “i _out ” and the output voltage is represented as “v _out ”.

The primary side switching elements SW1 and SW2 are connected between the primary coils 31 and 32 and the negative electrode of the power source Bt, and the reflux diodes are connected in parallel, and positive and negative voltages are alternately applied to the primary coils 31 and 32. It works alternately.
In the present embodiment, MOSFETs having body diodes are used as the primary side switching elements SW1 and SW2. The body diode in this configuration is interpreted as being included in the “freewheeling diode”. In the figure, the capacitance existing between the drain and the source is indicated by a symbol as a capacitance component. This capacity is a combined capacity of a junction capacity of a transistor and a diode and a capacitor connected in parallel. This capacitance symbol does not necessarily mean an independent capacitor element.

The primary side switching elements SW1 and SW2 have drains connected to the primary coils 31 and 32 and sources connected to the negative electrode of the power source Bt. A gate signal is input from the switching operation unit 18 to the gates of the primary side switching elements SW1 and SW2.
In other embodiments, a transistor such as an IGBT may be used as the switching element. In that case, the base, collector, and emitter of the transistor are interpreted as a gate equivalent electrode, a drain equivalent electrode, and a source equivalent electrode, respectively.

Next, the difference between the push-pull type DC / DC converter 102 having the configuration of the switching elements SW3 and SW4 as the secondary side rectifying elements and the diode configuration will be described with reference to FIG.
In the push-pull DC / DC converter 102, the secondary side switching elements SW3 and SW4 rectify the current flowing through the secondary coils 33 and 34. The secondary side switching elements SW3 and SW4 are also connected in parallel with freewheeling diodes.
In the present embodiment, MOSFETs having body diodes are used as the secondary side switching elements SW3 and SW4 as well as the primary side switching elements SW1 and SW2.

The secondary side switching elements SW 3 and SW 4 have drains connected to the secondary coils 33 and 34 and sources connected to the smoothing inductor 7. A gate signal is input from the switching operation unit 18 to the gates of the secondary side switching elements SW3 and SW4.
The switching operation unit 18 operates the secondary side switching elements SW3 and SW4 in addition to the primary side switching elements SW1 and SW2. Specifically, “synchronous rectification” is performed to turn on SW3 and SW4 in order to reduce loss due to energization (that is, conduction loss) during the energization period of SW3 and SW4.

Similar to the primary side switching elements SW1 and SW2, in other embodiments, transistors such as IGBTs may be used as the secondary side switching elements SW3 and SW4. In that case, the base, collector, and emitter of the transistor are interpreted as a gate equivalent electrode, a drain equivalent electrode, and a source equivalent electrode, respectively.
Hereinafter, the “switching element” will be omitted as appropriate, and simply referred to as “SW1 to SW4”.

Next, an outline of the operation of the push-pull DC / DC converters 101 and 102 will be described with reference to FIG.
Three sections are defined based on the ON / OFF states of the primary side switching elements SW1 and SW2. In section I, SW1 is on and SW2 is off. In section II, SW1 is off and SW2 is on. In section III, both SW1 and SW2 are off. In the push-pull type DC / DC converters 101 and 102, SW1 and SW2 are alternately turned on with a period in which both are turned off, as in the section I → III → II → III → I.
In section III, the magnetic energy stored in the smoothing inductor 7 decreases and the load current i _L decreases. On the other hand, in the sections I and II, magnetic energy is accumulated in the smoothing inductor 7 and the load current i _L increases.

In the case of the configuration in which the synchronous rectification is performed using the secondary side switching elements SW3 and SW4, SW3 is turned on after the transition from the section I to the section III, and is turned off after the section II is passed and before the transition from the section III to the section I is performed. To do. SW4 turns on after transitioning from section II to section III, passes through section I, and turns off before transitioning from section III to section II.
Here, the reason why the switching timing of the secondary side switching elements SW3 and SW4 is slightly shifted from the switching timing of the primary side switching elements SW1 and SW2 is to secure a dead time for preventing a short circuit.

Hereinafter, a valley region where the load current i _L changes from a decrease to an increase, that is, a region where the load current i _L shifts from the section III to the section I or the section II is referred to as a “load current minimum region”.
The minimum value of the load current i _L in the minimum region is determined by the timing of turning on the primary side switching elements SW1 and SW2. For example, if SW1 and SW2 are turned on while the drain-source voltage of SW1 and SW2 (hereinafter “DS-to-DS”) is still high, turn-on loss occurs. In addition, when SW1 and SW2 are turned on, there is a problem that a surge voltage is generated in the secondary diode due to resonance caused by various parasitic components.

The present embodiment relates to the operation in the minimum load current region of the push-pull type DC / DC converter, and is characterized in that the turn-on loss is reduced with a simple configuration using an excitation current.
Hereinafter, in describing the operation in the minimum load current region in detail, the “scene where the transition is made from the section III to the section I” circled in FIG. 3, that is, the scene where the primary side switching element SW1 is turned on will be described as a representative. . In the configuration in which synchronous rectification is performed on the secondary side, the primary side switching element SW1 is turned on after the secondary side switching element SW3 is turned off.

  Therefore, in the following description, the current and voltage changes of the secondary side switching element SW3 and the primary side switching element SW1 are mainly referred to. On the contrary, in the “scene where the transition is made from the section III to the section II”, it is possible to similarly interpret the secondary-side switching element SW4 and the primary-side switching element SW2 while appropriately setting the signs and the like of the current direction.

Next, with reference to FIGS. 4 to 9, the operation in the minimum load current region will be described for the push-pull DC / DC converter having two configurations. For details, refer to the time chart of FIG. 4 and the current path diagrams of FIGS. 5 and 6 for the configuration in which the secondary rectifier element is a diode or a switching element and the synchronous rectification control is stopped before the reverse rotation start time. . 5 and 6 illustrate a configuration in which the secondary rectifying elements are switching elements SW3 and SW4. In addition, referring to the time chart of FIG. 7 and the current path diagram of FIG. 8 for the configuration in which the secondary rectifier element is a switching element and the synchronous rectification control is continued after the reverse rotation start time.
In the following figures, physical quantities such as current, voltage, and inductance used in mathematical expressions described later are written in italics. On the other hand, “Ton, Toff, Tsw, Tadd”, which are time variables, are described in their true form in the figure. In addition, the subscript part of “on, off, sw, add” is described with a subscript in the mathematical expression and is written with a normal character in the text.

First, a configuration will be described in which the secondary side rectifying elements are the diodes DI3 and DI4 or the switching elements SW3 and SW4 and the synchronous rectification control is stopped before the reverse rotation start time tb.
In FIG. 4, in order from the top, the load current i _L and the excitation current i _Lm , the SW1 current i _sw1 , the SW1 voltage v _sw1 , the SW3 voltage v _sw3 , and the gate voltage of each switching element SW1 to SW4, that is, the switching operation An input change of the ON signal from the unit 18 is shown. Regarding the secondary side rectifying element, the gate voltage is always shown as 0 in the case of the configuration using the diodes DI3 and DI4. Further, in the case of a configuration in which synchronous rectification is performed using the secondary side switching elements SW3 and SW4, the gate voltage is indicated by a broken line.

Here, terms used in this specification are defined.
During the period in which the load current when all the primary side switching elements SW1 and SW2 are off flows back through the secondary side rectifying elements SW3 and SW4, the secondary side rectifying elements return via the load Ld and the smoothing inductor 7. Assuming that the current is an “element load current”, the excitation current i _Lm flowing in the excitation inductor of the transformer 20 is assumed to have no change in the excitation current in the section III with respect to the excitation current value accumulated in the sections I and II. The excitation current i _Lm converted to the current flowing per secondary side rectifying element is referred to as “converted excitation current”.
Also, the state where the absolute value of the element load current is equal to or greater than the absolute value of the converted excitation current is defined as `` normal state '', and the state where the absolute value of the element load current is lower than the absolute value of the converted excitation current Is defined. Attention is paid to the magnitude relationship between the element load current and the converted excitation current.

Unless otherwise specified in the description of the present embodiment, the secondary side rectifying element includes two elements having the same electrical specifications, and the winding ratio between the transformer primary side and the secondary side is 1: 1. . In this case, the element load current is represented by (i _L / 2), and the converted excitation current is represented by (i _Lm / 2).
That is, when comparing the load current corresponding to the element and the converted excitation current, the currents _½ of the load current i _L and the excitation current i _Lm are compared with each other, so the load current i _L and the excitation current i _Lm are compared. Even if they are compared, the judgment of the magnitude relationship is the same. Therefore, FIG. 4 shows the magnitude relationship between the load current i _L and the excitation current i _Lm .

In FIG. 4, five modes (a), (b), (d), (e), and (f) are defined in the minimum load current region. Note that mode (c) does not exist.
5 and 6, the load current i _L in each mode is indicated by a solid line, and the excitation current i _Lm is indicated by a broken line. Also, the sign of each current and voltage is defined as follows.

The load current i _L has a positive direction from the secondary center tap CT2 of the transformer to the secondary side switching elements SW3 and SW4 via the smoothing inductor 7.
The exciting current i _Lm has a positive direction in the direction of the broken-line arrow in each figure.
For the SW1 current i _{sw1 and the} like flowing through the switching element, the forward direction from the drain to the source, that is, the downward direction in the figure is the positive direction. The current in the negative direction flows in the direction opposite to the arrow in the figure, that is, upward in the figure.
The SW1 voltage v _sw1 , SW3 voltage v _sw3, etc. mean the DS voltage of the switching element, and as shown by the arrows in the figure, the drain potential with respect to the source potential is defined as a positive voltage. The SW2 voltage v _sw2 is complementary to the SW1 voltage v _sw1 and is indicated by a broken line.

Subsequently, each mode will be described in order.
In mode (a), the primary side switching element SW2 is on and the switching element SW1 is off. At this time, the load current i _L is in a normal state where the excitation current i _Lm is _{equal to} or greater than the excitation current i _Lm , and the load current i _L gradually increases. In the case of a configuration in which synchronous rectification is performed on the secondary side, the secondary side switching element SW3 is turned on as indicated by a broken line.
In mode (a), the load current i _L and the excitation current i _Lm flow to the primary side via the switching element SW2. In addition, a load current i _L flows through the switching element SW3 or the diode DI3 on the secondary side.

When the primary side switching element SW2 is turned off at the maximum load current time ta, the mode shifts to the mode (b), and the load current i _L changes from increasing to decreasing. In mode (b), both the primary side switching elements SW1 and SW2 are in the OFF state, and the load current i _L gradually decreases, but is in a normal state range _{equal to} or greater than the excitation current i _Lm .
In the configuration in which synchronous rectification is performed by the secondary side switching elements SW3 and SW4, SW4 is turned on when the mode (b) is entered. Further, SW3 is turned off before the reverse rotation start time tb.
In mode (b), the load current i _L and the excitation current i _Lm circulate via the secondary side switching elements SW3 and SW4 or the diodes DI3 and DI4 and do not flow to the primary side.

When the reverse rotation start time tb when the load current i _L becomes equal to the excitation current i _Lm is passed, the mode shifts to the reverse mode (d). At the time of shifting to the mode (d), both the primary side switching elements SW1 and SW2 remain off. Further, in the case of a configuration in which synchronous rectification is performed on the secondary side, the switching element SW3 is already turned off.

Here, the terms “diode conduction impossible current i”, “primary side commutation current i _p ”, and “secondary side residual current i _s ” used in the following description will be described with reference to FIG. 9.
In FIG. 9A, symbols of the primary transformer voltage v1 and the secondary transformer voltage v2 are added based on the current path diagram of the mode (d) in FIG. The primary-side transformer voltage v1 is a voltage that represents the potential at one end of the primary coil 31 that is not on the center tap CT1 side with reference to the potential at one end of the primary coil 32 that is not on the center tap CT1 side. The secondary-side transformer voltage v2 is a voltage that represents the potential at one end of the secondary coil 33 that is not on the center tap CT2 side, with reference to the potential at one end that is not on the center tap CT2 side of the secondary coil 33.

The secondary side switching elements SW3 and SW4 may be replaced with diodes DI3 and DI4. The primary side switching elements SW1 and SW2, and the secondary side switching elements SW3 and SW4 or the secondary side diodes DI3 and DI4 are premised on a configuration in which two elements having the same electrical specifications are provided in parallel. .
FIG. 9B is a circuit model showing an equivalent circuit of FIG. 9A, that is, the mode (d) of FIG. i is a diode conduction impossible current obtained by subtracting the converted excitation current (i _Lm / 2) from the element load current (i _L / 2). i _p is the primary side commutation current, i _s denotes a secondary residual current.
C _ds1, C _ds2, C _ds3 is a DS capacitance of the switching elements.

When the load current i _L of the secondary side falls below the exciting current i _Lm, the exciting current i _Lm will be unable to conduct the body diode of the secondary-side switching element SW3. Therefore, the diode conduction impossible current i obtained by subtracting the converted excitation current (i _Lm / 2) from the element current load current (i _L / 2) is multiplied by the impedance ratio of the primary side switching element path and the secondary side switching element path. The primary commutation current i _p thus commutated from the secondary side to the primary side.

To flow the charge in the capacitor present in parallel with the primary side commutation current i _p is the primary side switching element SW1 in the direction to charge the capacitor that exists in parallel with the switching element SW2, SW1 current i _sw1 is negative. By commutating current i _p flows, SW1 voltage v _sw1 decreases gradually, SW2 voltage v _sw2 gradually increases. Therefore, in the configuration of FIG. 4, the voltage drop of the SW1 voltage v _sw1 starts at the reverse rotation start time tb. Further, when a current obtained by subtracting the primary side commutation current i _p from diode conduction allowed current i and the secondary side residual current i _s, for charging the capacitor that exists in parallel to SW3 voltage by the secondary-side residual current i _s The DS voltage v _sw3 of the secondary side switching element SW3 rises.

The SW1 voltage v _sw1 drops to zero voltage at the zero voltage arrival time td.
In mode (e), which is the period of the reverse state after the zero voltage arrival time td, the voltages of SW1, SW2, and SW3 have reached the steady voltage of section I, and therefore charging / discharging of the capacitors present in parallel with each SW Current stops flowing. All of the exciting current i _Lm corresponding to the difference from the load current i _L flows through the primary side and regenerates to the power source Bt.
Regardless of whether the primary side switching element SW1 is turned on or off, the SW1 voltage v _sw1 is zero voltage, so that a voltage in the direction opposite to that in the section II is applied to the coil 31 of the transformer 20. For this reason, the SW1 current i _sw1 increases in the forward direction.

As shown in FIGS. 1 and 2, the switching time calculation unit 13 acquires information such as the input voltage v _in , the output voltage v _out , the output current i _out and the load current i _L , and based on these information, “Theoretical on time Ton” and “theoretical off time Toff” described later are calculated. Assuming that there is no error in the theoretical off time Toff, the zero voltage arrival time td corresponds to the timing at which the theoretical off time Toff has elapsed.
Based on the theoretical on-time Ton and the theoretical off-time Toff calculated by the switching time calculation unit 13, the switching operation unit 18 performs “next” during the reverse rotation state, that is, during the mode (d) or the mode (e). SW1 which is the “primary side switching element in the turn-on order” is turned on. A specific configuration for calculating the theoretical off time Toff will be described later.

Preferably, the switching time calculation unit 13 calculates “adjusted off time Toff ^* ” obtained by adding the “extension time Tadd” to the theoretical off time Toff, which is less than the mode (e) period. When the elapsed timing of the theoretical off time Toff ideally matches the zero voltage arrival time td, the timing at which the adjusted off time Toff ^* has elapsed is within the period of the mode (e).
Accordingly, the switching operation unit 18 turns on the primary side switching element SW1 at the timing when the adjustment off time Toff ^* has elapsed, so that the “zero voltage switching” can be reliably realized when the SW1 voltage v _sw1 is in the zero voltage state. it can.

During the period from the zero voltage arrival time td to when the primary side switching element SW1 is turned on, a circuit operation is performed in which the load current i _L increases in the same manner as when the switching element SW1 is off, even though the switching element SW1 is off. . That is, this period is equivalent to the ON period of the primary side switching element Sw1 in circuit operation, and it is considered that zero voltage switching can be realized if the primary side switching element SW1 is turned on within this period. Therefore, by adding the extension time Tadd to the theoretical off-time Toff, it is possible to ensure robustness against deviations and variations in the operation timing of the switching elements.

In addition, if there is a possibility that the elapsed timing of the theoretical off time Toff may deviate from the zero voltage arrival time td due to the calculation error of the theoretical off time Toff, the elapsed timing of the adjustment off time Toff ^* is considered in consideration of the error. It is preferable to set the extension time Tadd appropriately so as to be within the period of the mode (e).
Furthermore, the dead time DT, which is the time from the turn-off of the secondary side switching element SW3 to the turn-on of the primary side switching element SW1, is set so as to ensure a minimum time due to the characteristics of the element.

When the SW1 current i _sw1 increases in the forward direction in the mode (e) and the normal state return time te that is equal to the current value flowing through the power source Bt on the primary side is passed, the mode (f) is shifted to the normal state. In mode (f), the primary side switching element SW1 is in the ON state, and the SW1 current i _sw1 flows in the forward direction. Thereafter, the positive SW1 current i _sw1 gradually increases.

Next, the operation in the minimum load current region will be described for the configuration in which the secondary rectifier elements are the switching elements SW3 and SW4 and the synchronous rectification control is continued after the reverse rotation start time tb.
Regarding the operation of the secondary side switching element SW3, in the operation indicated by the broken line in FIG. 4 of the above configuration, the secondary side switching element SW3 is turned off during the mode (b) from the load current maximum time ta to the reverse rotation start time tb. Is done.
On the other hand, in the configuration of FIG. 7, the secondary side switching element SW3 is maintained in the ON state for the duration T _cont after the reverse rotation start time tb and continues the synchronous rectification. The state during the duration time T _cont is set to a mode (c) that does not exist in FIG.

FIG. 8 shows current paths in modes (a), (b), and (c) in the synchronous rectification operation. The mode (a) and the mode (b) in FIG. 8 are different from the mode (a) and the mode (b) in FIG. 5 in that the current passes between the DSs of the secondary side switching element SW3. Therefore, in FIG. 8, it has described so that the arrow which shows an electric current path may pass between DS.
Even when the load current i _L falls below the excitation current i _Lm in the mode (c), the excitation current i _Lm can flow through the DS of the secondary-side switching element SW3 in the on state. Therefore, the current path in mode (c) is the same as in mode (b).

If the secondary side switching element SW3 is turned off at the synchronous rectification stop time tc when the duration T _cont has elapsed from the reverse rotation start time tb, the exciting current i _Lm cannot conduct the switching element SW3. Therefore, the impedance ratio between the primary side switching element path and the secondary side switching element path in the diode conduction impossible current i obtained by subtracting the converted excitation current (i _Lm / 2) from the element current load current (i _L / 2) the primary side commutation current i _p multiplied is commutated to the primary side from the secondary side. And it transfers to mode (d).

As described above, in the configuration in which the synchronous rectification control is continued, the voltage drop of the SW1 voltage v _sw1 starts from the synchronous rectification stop time tc. In addition, the initial value of the diode conduction impossible current i at the start of the voltage drop is not zero, and the initial current value increases in proportion to the duration T _cont .
In the mode (d), the SW1 voltage v _sw1 drops in a slope shape by the primary-side commutation current i _p and becomes zero at the zero voltage arrival time td. Further, while the SW1 voltage v _sw1 decreases, the voltage v _sw3 of the secondary side switching element SW3 increases.

  The relationship between the elapsed time of the theoretical off time Toff in FIG. 7 and the zero voltage arrival time td, the setting of the extension time Tadd, and the current and voltage changes in the modes (e) and (f) are described with reference to FIG. It is the same. Further, the current paths in the modes (d), (e), and (f) are shown in the same manner as in FIG. Strictly speaking, it differs from FIG. 6 in that the reflux current can pass between the DSs of the secondary-side switching element SW4 in the on state, but the illustration is omitted because it does not affect the overall operation.

As described above, in the present embodiment, based on the theoretical on-time Ton and the theoretical off-time Toff calculated by the switching time calculation unit 13, the switching operation unit 18 “primary in the next turn-on order” during the reverse rotation state. The "side switching element" is turned on. Thereby, an appropriate switching timing for reducing the turn-on loss can be determined with a simple configuration.
Next, a specific configuration of the switching time calculation unit 13 will be described as the first and second embodiments. Hereinafter, the code | symbol of the switching time calculation part of each embodiment shall be "131,132".

(First embodiment)
The configuration of the switching time calculation unit 131 of the first embodiment will be described with reference to FIG. As shown in FIG. 10A, the switching time calculation unit 131 includes a duty control unit 14, a theoretical off time calculation (“Toff calculation” in the drawing) 151, and an adjustment off time calculation (“Toff ^* calculation in the drawing). ]) Part 16 is included.
The duty control unit 14 acquires at least one of the output current i _{out and the} output voltage v _out and calculates the theoretical on-time Ton by feedback control on these. A load current i _L may be used instead of the output current i _out .

Here, with reference to FIG. 11, the well-known duty control in which the switching period Tsw is set as a fixed value will be described. The duty control unit 14 calculates a duty that is a ratio of the on-time Ton to the switching cycle Tsw by PI control or the like so that the output current i _out or the output voltage v _out matches the target value, and outputs the duty to the switching operation unit 18. To do. The switching operation unit 18 compares the duty with the carrier and generates gate signals of the primary side switching elements SW1 and SW2 by PWM modulation. The off time Toff is calculated by subtracting the on time Ton from the switching period Tsw that is a fixed value.
The switching time calculation unit 131 of the first embodiment is the same as the known technique in FIG. 11 in that the duty control unit 14 calculates the duty and outputs it as information on the on time Ton. However, the switching period Tsw is not a fixed value, and is different from the known technique in that it calculates the off time Toff in addition to the on time Ton.

Next, with reference to FIG. 3 and FIG. 10B, “theoretical on time Ton” and “theoretical off time Toff” are defined. FIG. 10B shows the “element current load current (i _L / 2)”, but the part other than the part discussed in the text by comparing the absolute current value with the converted excitation current (i _Lm / 2). Here, it is simply written as “load current i _L ”. The same applies to FIGS. 13 and 14 described later.
As described above, in FIG. 3, the load current i _L increases in the section I and the section II in which either the primary side switching element SW1 or SW2 is on, and the primary side switching elements SW1 and SW2 are both off. In III, the load current i _L decreases.
As shown in FIG. 10B, the time when the load current i _L theoretically increases is called a theoretical on-time Ton, and the time when the load current i _L theoretically decreases is called a theoretical off-time Toff. The sum of the theoretical on-time Ton and the theoretical off-time Toff is the switching period Tsw.

  The switching period Tsw corresponds to the sum of the section I or the section II and the section III in FIG. That is, the switching cycle Tsw defined here is not the ON / OFF cycle of each of the primary side switching elements SW1 and SW2, but “the ON time of one of the primary side switching elements” and “all the primary side switching elements. It is the total of “the time when both elements are turned off”.

In the present embodiment, it is preferable to output the adjusted off time Toff ^* obtained by adding the extended time Tadd, instead of outputting the calculated theoretical off time Toff as it is. Therefore, in order to distinguish from this adjusted off time Toff ^* , “theory” is added as a prefix to the theoretically calculated off time. In consideration of the balance with the “theoretical off time Toff”, the theoretically calculated on time is also referred to as “theoretical on time Ton”.

The increase amount Δi of the load current i _L during the theoretical on-time Ton and the decrease amount Δi of the load current i _L during the theoretical off-time Toff use the input voltage v _in , the output voltage v _out , and the inductance L of the smoothing inductor 7. And expressed by the equations (1.1) and (1.2).

In the description of this part, the load current i _{L is} described as “element load current (i _L / 2)”. When the peak value of the element current load current (i _L / 2) at the timing when the primary side switching element SW1 or SW2 is turned off and starts to decrease is assumed to be the initial value, the element current load current (i _L / 2) changes. The amount Δi is equal to the difference between the initial value of the element current load current (i _L / 2) and the converted excitation current (i _LM / 2). That is, when the transformer turns ratio is 1: N, Equation (2.1) is established. In the text, N = 1 is described.
Moreover, Formula (2.2) is obtained from Formula (1.2) and Formula (2.1).

The theoretical off time calculation unit 151 of the first embodiment uses the equation (2.2) to reduce the element load current (i _L / 2) from the initial value to the converted excitation current (i _LM / 2). Is calculated as the theoretical off time Toff.
At this time, the theoretical off-time calculation unit 151 obtains the load current i _L and the output voltage v _out that are already detected as information of the duty control unit 14 in an overlapping manner. However, since it is not easy to detect the excitation current i _LM , an estimated value estimated using a circuit constant or the like may be used.

Subsequently, the adjustment off-time calculating unit 16 calculates an adjustment off-time Toff ^* obtained by adding the extension time Tadd to the theoretical off-time Toff by the equation (3), and outputs it to the switching operation unit 18. This technical significance will be described together in the second embodiment.

Thus, the theoretical time ON Ton and the adjustment OFF time Toff ^* are output from the switching time calculator 131 to the switching operation unit 18. The switching time calculation unit 131 calculates the switching cycle Tsw by adding the theoretical on-time Ton and the adjustment off-time Toff ^*, and outputs the switching cycle Tsw and the theoretical on-time Ton to the switching operation unit 18. It may be.

(Second Embodiment)
The configuration of the switching time calculation unit 132 of the second embodiment will be described with reference to FIG. The switching time calculation unit 132 includes a theoretical off time calculation unit 152 having a configuration different from that of the theoretical off time calculation unit 151 of the first embodiment. Further, the theoretical on-time Ton calculated by the duty control unit 14 is directly output to the switching operation unit 18 and is also output to the theoretical off-time calculation unit 152.

From the above equations (1.1) and (1.2), equation (4) is obtained. Theory off time calculator 152 of the second embodiment obtains the theoretical ON time Ton, the input voltage v _in and the output voltage v _out, using Equation (4) to calculate the theoretical off time Toff.

As in the first embodiment, the adjustment off time calculation unit 16 calculates the adjustment off time Toff ^* by adding the extension time Tadd to the theoretical off time Toff. Then, the switching time calculation unit 132 outputs the theoretical on time Ton and the adjustment off time Toff ^* to the switching operation unit 18.

Next, the technical significance of the extended time Tadd will be described with reference to FIGS.
As shown in FIG. 13, when the extension time Tadd is not set, the switching cycle Tsw is constant in the steady state, and the valley value and the peak value of the load current i _L repeat the same value. At this time, the operation is basically performed in the region of “element current load current (i _L / 2)> converted excitation current (i _Lm / 2)”, and the reverse rotation state does not occur.

In contrast, FIG. 14 shows the behavior when the extension time Tadd is set.
Under the condition of “element current load current (i _L / 2)> converted excitation current (i _Lm / 2)” shown in (a), the adjustment off time Toff ^* longer than the theoretical off time Toff is used. The load current decrease amount Δi (Toff) is larger than the ON time increase amount Δi (Ton). As a result, the trough value and peak value of the element current load current (i _L / 2) gradually decrease.

After the valley value of the element load current (i _L / 2) falls below the converted excitation current (i _Lm / 2), as shown in (b), “element load current (i _L / 2) <converted excitation The switching elements SW1 and SW2 are turned on during the condition of “current (i _Lm / 2)”, that is, during the reverse rotation state.
Then, the theoretical on-time Ton is increased by the feedback control of the output current i _out or the output voltage v _out by the duty control unit 14. Further, as the theoretical on-time Ton increases, the theoretical off-time Toff calculated by the theoretical off-time calculator 152 according to Equation (4) also increases. Therefore, the switching cycle Tsw is variable.

As a result, the increase amount Δi (Ton) of the on time in each cycle becomes equal to the load current decrease amount Δi (Toff) of the off time, and the operation is stabilized.
In other words, the switching operation unit 18 turns on the primary side switching element in the turn-on order when the adjustment off time Toff ^* in which the extension time Tadd is added every cycle has elapsed, thereby turning on in the reverse state. It can be performed stably.

4 and 7, the timing at which the theoretical off time Toff has elapsed corresponds to the zero voltage arrival time td at which the mode (d) shifts to the mode (e). However, in reality, when the error of the theoretical off time Toff is taken into consideration, it is considered that the timing at which the theoretical off time Toff has elapsed is included in the mode (d) or the mode (e) in the reverse state.
When the switching operation unit 18 turns on the primary side switching element SW1 in the mode (d), the zero voltage switching is not realized because the SW1 voltage v _sw1 is not lowered to 0.

Therefore, when the extension time Tadd that is equal to or greater than the error assumed for the theoretical off time Toff is added to the theoretical off time Toff, the timing at which the adjustment off time Toff ^* has elapsed is later than the zero voltage arrival time td. Further, by setting the extension time Tadd so that the adjustment off time Toff ^* does not exceed the normal state return time te, the primary side switching element SW1 is always turned on in the mode (e). Therefore, zero voltage switching can be realized with certainty.
Note that when the extension time Tadd is set in the first embodiment, there is no effect that the theoretical off time Toff increases as the theoretical on time Ton increases. However, the same effect as in the second embodiment can be obtained in that the turn-on timing can be prevented from being earlier than the zero voltage arrival time td due to an error in the theoretical off time Toff.

(effect)
The effects of this embodiment including the first and second embodiments will be described. Here, a case where the primary side switching element SW1 is turned on will be described as an example.
(1) The switching operation unit 18 turns on the primary side switching element SW1 during the reverse rotation state based on the theoretical on-time Ton and the theoretical off-time Toff calculated by the switching time calculation units 131 and 132. Even if the error of the theoretical off time Toff is taken into consideration, the primary side switching element SW1 is turned on at least in the mode (d) or the mode (e) in the reverse rotation state.

First, even when the primary side switching element SW1 is turned on in the mode (d), it is turned on after the SW1 voltage v _sw1 starts to decrease, so that it is compared with the case where the primary side switching element SW1 is turned on during the period of the mode (b) or mode (c). , Turn-on loss can be reduced. Further, when the primary side switching element SW1 is turned on in the mode (e) after the zero voltage arrival time td, zero voltage switching can be realized, and the turn-on loss can be particularly reduced.
Thus, in the present embodiment, the turn-on loss of the primary side switching element SW1 is reduced with a simple configuration using the commutation action of the excitation current i _Lm without providing an auxiliary switch or the like as in Patent Document 1. be able to.

(2) Further, in the present embodiment, the switching time calculation units 131 and 132 calculate the adjustment off time Toff ^* obtained by adding the extension time Tadd to the theoretical off time Toff, and the switching operation unit 18 passes the adjustment off time Toff ^*. When this occurs, the primary side switching element SW1 is turned on. By appropriately setting the extension time Tadd, zero voltage switching can be reliably realized. Therefore, robustness can be ensured with respect to deviations and variations in the operation timing of the switching elements.

In the second embodiment, as a result of setting the extension time Tadd, the theoretical off time calculating unit 152 calculates the theoretical off time 152 as the theoretical on time Ton increases due to feedback control of the output current i _out or the output voltage v _out. Time Toff increases. As a result, the increase amount Δi (Ton) of the on-time in each cycle becomes equal to the load current decrease amount Δi (Toff) of the off-time, and the operation is stabilized.

Referring to FIG. 15 and FIG. 16, the behavior when the push-pull type DC / DC converter is operated in the normal continuous mode and the zero voltage switching mode will be compared. In each figure, scale values on the horizontal and vertical axes are not shown. However, the scale of (a) continuous mode and (b) zero voltage switching mode shall be the same.
FIG. 15 shows the DS voltage v _sw1 and current i _sw1 of the primary side switching element SW1. In the continuous mode of the comparative example, the SW1 is turned on and the current i _sw1 flows before the DS voltage v _sw1 decreases to 0, so that a turn-on loss occurs due to the product of the current i _sw1 and the voltage v _sw1 . In contrast, in the zero voltage switching mode, the SW1 is turned on after the inter-DS voltage v _{sw1 drops} to 0, so that the turn-on loss can be reduced to zero.

FIG. 16 shows inter-DS voltages v _sw1 and v _sw3 of the primary side switching element SW1 and the secondary side switching element SW3. In the continuous mode of the comparative example, the voltage drop rate (dv _sw1 / dt) of the primary side switching element SW1 is relatively large, and the surge voltage of the secondary side switching element SW3 becomes large. On the other hand, in the zero voltage switching mode, the voltage drop rate (dv _sw1 / dt) of the primary side switching element SW1 can be reduced, and the surge voltage of the secondary side switching element SW3 can be reduced.

(3) The information used by the switching time calculation units 131 and 132 of the first and second embodiments for calculating the theoretical off time Toff is for controlling the basic operation of the push-pull DC / DC converters 101 and 102. It is necessary information originally. Therefore, by determining the reverse rotation state based on the theoretical off time calculated based on these pieces of information, it is not necessary to add a sensor dedicated to detection, and it is not necessary to consider the influence of sensor accuracy.
In particular, in the second embodiment, since it is not necessary to estimate the excitation current i _Lm as in the first embodiment, the theoretical off time Toff can be calculated with higher accuracy.

(Other embodiments)
In the above embodiment, in order to reliably realize zero voltage switching, the switching operation unit 18 turns on the primary-side switching element SW1 when the adjustment off time Toff ^* obtained by adding the extension time Tadd to the theoretical off time Toff has elapsed. . However, for example, when a program for adjusting the duty so that the load current IL at the end of the theoretical off time Toff is always zero is constructed, and the theoretical off time Toff has passed without adding the extension time Tadd, the primary side The switching element SW1 may be turned on.

In the first embodiment, the element current load current (i _L / 2) and the converted excitation current (i _Lm / 2) are obtained by the expression (2.2) obtained from the expressions (1.2) and (2.1). ), And the theoretical off time Toff is calculated based on the output voltage _Vout . Similarly, the theoretical on-time Ton may be calculated from the equation (5) obtained from the equations (1.1) and (2.1).

In the above description, since the winding ratio between the transformer primary side and the secondary side is 1: 1 in a configuration in which two secondary side rectifying elements are connected in parallel, the total excitation current i _{Lm is set} to 1. Then, the converted excitation current flowing per secondary side rectifier element is (i _Lm / 2).
On the other hand, when two secondary rectifying elements are connected in parallel and the winding ratio is 1: N, the converted excitation current is (i _Lm / 2N). In general, in a configuration in which M secondary side rectifying elements are connected in parallel, the converted excitation current is (1 / (M × N)).

In the time charts of FIGS. 4 and 7, the positive and negative directions are set so that the load current i _L and the excitation current i _Lm can be directly compared without reversing the signs. However, depending on the definition of the current direction, it may be necessary to invert the sign as appropriate for comparison. Generally speaking, the determination of “normal state” and “reversed state” is performed based on a comparison between “absolute value of element load current” and “absolute value of converted excitation current”.

In the present invention, the specific types of the power source Bt and the load Ld or the specific numerical ranges of the input voltage v _in and the output voltage v _out are not limited.
As mentioned above, this invention is not limited to the said embodiment at all, In the range which does not deviate from the meaning of invention, it can implement with a various form.

101, 102... Push-pull type DC / DC converter,
13 (131, 132) ... switching time calculation unit,
18 ... switching operation part,
20 ... Transformer,
31, 32 ... primary coil,
33, 34 ... secondary coil 7 ... smoothing inductor,
SW1, SW2 ... primary side switching elements,
DI3, DI4 ... Diode (secondary rectifier),
SW3, SW4... Secondary side switching element (secondary side rectifying element).

Claims (3)

A push-pull type DC / DC converter connected between a power source (Bt) and a load (Ld) for converting DC power,
A plurality of primary coils (31, 32) and a plurality of secondary coils (33, 34) constituting an exciting inductor of the transformer (20);
A plurality of primary-side switching elements (SW1, SW2) connected between the plurality of primary coils and the power source and connected in parallel with freewheeling diodes;
A plurality of secondary side rectifier elements (DI3, DI4, SW3, SW4) connected to the plurality of secondary coils and capable of rectifying the current flowing through the secondary coil;
A smoothing inductor (7) connected between the secondary-side rectifying element and the load;
A theoretical on-time (Ton) in which any of the primary-side switching elements is turned on and a load current (i _L ) flowing through the load and the smoothing inductor is theoretically increased; and all the primary-side switching elements A switching time calculation unit (13) for calculating a theoretical off time (Toff), which is a time during which the load current is theoretically reduced after turning off,
The primary side switching element is operated based on the theoretical on time and the theoretical off time calculated by the switching time calculation unit, and the secondary side switching element is further operated when the secondary side rectifying element is a switching element. A switching operation unit (18);
With
In the period when the load current when all of the primary side switching elements are off flows back through the secondary side rectifier element, the load current that flows back per one secondary side rectifier element is expressed as the element current load current (i _L / 2), and the excitation current converted to the current flowing per one of the secondary side rectifiers is the converted excitation current (i _Lm / 2N).
The switching operation unit is
After a transition from a normal state where the absolute value of the element current load current is greater than or equal to the absolute value of the converted excitation current to a reverse state where the absolute value of the element current load current is lower than the absolute value of the converted excitation current, the reverse rotation During the state, turn on the primary-side switching elements in the next turn-on order ,
In the configuration in which the theoretical off time is Toff, the peak value of the element load current is (i _L / 2), and the winding ratio between the primary side and the secondary side of the transformer is 1: N, the primary side switching element is The converted excitation current when all are off is expressed as (i _Lm / 2N), the output voltage output to the load when all the primary side switching elements are off is expressed as v _out , and the inductance of the smoothing inductor is expressed as L.
The switching time calculation unit is a push-pull DC / DC converter that calculates the theoretical off-time according to the following equation .

The switching operation unit is
2. The push-pull DC / DC converter according to claim 1 , wherein after the drain-source voltage of the primary side switching element drops and reaches zero voltage, the primary side switching elements in the turn-on order are turned on. The switching time calculation unit further calculates an adjustment off time (Toff ^* ) obtained by adding a predetermined extension time (Tadd) to the theoretical off time,
The switching operation unit is
3. The push-pull DC / DC converter according to claim 2 , wherein when the adjustment off time has elapsed, the primary side switching elements in the turn-on order are turned on.

Images