JP3669953B2 - Switching power supply - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a switching power supply device for supplying a DC stabilized voltage to industrial or consumer electronic devices.
[0002]
[Prior art]
A power source in an electronic device such as a television, a VTR, or a personal computer must stably supply a constant DC voltage. A switching power supply device is preferable as such a power supply. The switching power supply device uses semiconductor elements such as MOSFETs, IGBTs, thyristors or the like as switches, and once converts the input DC voltage into AC voltage by turning on / off these switches. The converted AC voltage is converted into a stable DC voltage through a transformer, a rectifier circuit, and a smoothing circuit in order, and is output. In the switching power supply device, the ratio of the output voltage to the input voltage (voltage conversion rate) is substantially determined by the on / off time ratio of the switch. Therefore, the switching power supply device can control the on / off time ratio by controlling the switch, thereby stabilizing the output DC voltage.
[0003]
Since the power loss (switching loss) due to switching of the switch is generally small, the switching power supply device can supply power with high efficiency. Therefore, the switching power supply device is excellent in terms of energy saving.
The voltage conversion rate of the switching power supply device substantially depends only on the on / off time ratio of the switch, and does not substantially depend on the on / off switching frequency (switching frequency). On the other hand, in the switching power supply device, the higher the switching frequency, the smaller the reactance elements such as a transformer, an inductor, and a capacitor can be maintained while maintaining their respective performances. Therefore, the switching power supply device can be reduced in size and weight relatively easily while maintaining a constant output voltage.
[0004]
An example of a conventional switching power supply device that performs hard switching will be described below. FIG. 6 is a circuit diagram of a conventional switching power supply apparatus 100.
In the switching power supply device 100, the four switch units 101H, 101L, 102H, and 102L on the primary side of the transformer 3 form a full bridge. The switch portions 101H and 102H connected to the high-potential side input terminal 1a of the full bridge are referred to as high-side switch portions, and the switch portions 101L and 102L connected to the low-potential side input terminal 1b are referred to as low-side switch portions. Each switch part includes switch elements 1HS, 2HS, 3HS, and 4HS. These switch elements are semiconductor elements, for example, IGBTs. Each switch unit includes parasitic capacitors 1HC, 2HC, 3HC, and 4HC in parallel with the switch element. Each switch unit is controlled to be turned on / off by a switching control circuit 70.
[0005]
FIG. 7 is a waveform diagram of currents and voltages generated in each part of the circuit shown in FIG. 6 due to hard switching by the switching control circuit. Here, the current and voltage of each circuit portion are positive in the direction of the arrow shown in FIG.
The switching control circuit 70 outputs switching signals G1, G2, G3, and G4 to the switch elements 1HS, 1LS, 2HS, and 2LS, respectively. The switching signals G1, G2, G3, and G4 are each rectangular waves. The switch element is turned on while the switching signal shows a high potential (H), and the switch element is turned off while the switching signal shows a low potential (L).
[0006]
The switching control circuit 70 performs hard switching with respect to ON / OFF of the four switch units. Here, hard switching refers to switching that simultaneously turns on / off the high-side switch unit and the low-side switch unit. The hard switching by the switching control circuit 70 realizes the following three periods alternately at a predetermined time and in a predetermined cycle:
(1) The first period corresponds to periods T0 to T1 in FIG. In the first period, the first high-side switch unit 101H and the second low-side switch unit 102L are turned on, and the second high-side switch unit 102H and the first low-side switch unit 101L are turned off.
(2) The second period corresponds to periods T2 to T3 in FIG. In the second period, the first high-side switch unit 101H and the second low-side switch unit 102L are turned off, and the second high-side switch unit 102H and the first low-side switch unit 101L are turned on.
(3) The third period corresponds to the period T1 to T2 and the period T3 to T4 in FIG. 7, and is realized between the first period and the second period. In the third period, all four switch parts are turned off.
[0007]
Hereinafter, hard switching by the switching control circuit 70 will be described in order of time from time T0 to time T4 shown in FIG.
<Period T0-T1>
At time T0, the switching control circuit 70 simultaneously changes the first switching signal G1 and the fourth switching signal G4 from L to H, and simultaneously switches the first high-side switch unit 101H and the second low-side switch unit 102L. Turn on. On the other hand, the first low-side switch unit 101L and the second high-side switch unit 102H are both turned off.
[0008]
During the period T0 to T1, a substantially constant and positive input voltage Vin is applied to the primary winding 3a of the transformer 3 through the first high-side switch unit 101H and the second low-side switch unit 102L. Accordingly, the voltage across the primary winding 3a, that is, the primary voltage Vt is substantially equal to the input voltage Vin. Further, the primary current It of the transformer 3 flows through the primary winding 3a from the first connection point P to the second connection point Q, that is, in the direction of the arrow shown in FIG. At that time, a positive voltage Vin / n is induced in the secondary windings 3b and 3c of the transformer 3, respectively. Here, the turns ratio of the primary winding 3a, the first secondary winding 3b, and the second secondary winding 3c of the transformer 3 is assumed to be n: 1: 1 (n is a positive real number). Since the first rectifier diode 4b is on, the voltage V5 across the smoothing inductor 5 is substantially equal to Vin / n−Vout. Here, the voltage across the smoothing capacitor 6, that is, the output voltage Vout is positive. Since the smoothing capacitor 6 has a sufficiently large capacity, the output voltage Vout may be substantially constant. Therefore, the current I5 flowing through the smoothing inductor 5 increases linearly in the direction of the arrow shown in FIG. 6 during the period T0 to T1. However, since the inductance of the smoothing inductor 5 is sufficiently large, the increase in the current I5 is gradual. The voltage Vc across the second rectifier diode 4c is substantially equal to + 2Vin / n when the direction of the arrow shown in FIG. Therefore, the second rectifier diode 4c is off. Therefore, the current I5 of the smoothing inductor 5 is substantially equal to the current Ib flowing through the first rectifier diode 4b. As a result, in the period T0 to T1, the secondary current of the transformer 3 flows only through the first secondary winding 3b and increases linearly.
[0009]
The primary current It of the transformer 3 is equal to the sum of the exciting current for the transformer 3 and the primary side converted current determined by the secondary current of the transformer 3. As shown in FIG. 7, the primary voltage Vt maintains a substantially constant value Vin in the periods T0 to T1. Therefore, the excitation current included in the primary current It increases substantially linearly. On the other hand, since the secondary current of the transformer 3 increases linearly as already described, the primary side converted current increases linearly. Accordingly, the primary current It increases substantially linearly.
[0010]
<Period T1-T2>
At time T1, the switching control circuit 70 simultaneously changes the first switching signal G1 and the fourth switching signal G4 from H to L, and simultaneously turns off the first high-side switch unit 101H and the second low-side switch unit 102L. To do. As a result, all four switch portions are turned off, so that the input voltage Vin is not applied to the primary winding 3a of the transformer 3. That is, the primary voltage Vt of the transformer 3 rapidly drops from the input voltage Vin to 0. Therefore, the induced voltages of the secondary windings 3b and 3c of the transformer 3 are abruptly changed to zero. At that time, since the first rectifier diode 4b is already turned on, the voltage Vc across the second rectifier diode 4c suddenly changes to zero. Thereby, the second rectifier diode 4c is suddenly turned on. As a result, the secondary current of the transformer 3 starts to flow through both the first secondary winding 3b and the second secondary winding 3c. Thus, commutation occurs on the secondary side of the transformer 3 at time T1.
[0011]
In the period T1 to T2, no voltage is applied to the primary winding 3a of the transformer 3, and the primary voltage Vt is zero. Accordingly, since no voltage is induced in the secondary windings 3b and 3c of the transformer 3, the voltage V5 across the smoothing inductor 5 is substantially equal to the negative constant voltage −Vout. As a result, the current I5 of the smoothing inductor 5 decreases linearly with a gentle slope.
Since all the four switch units are turned off at time T1, the primary current It of the transformer 3 rapidly decreases to zero. Accordingly, the secondary current of the transformer 3 flows through the two secondary windings 3b and 3c in the opposite directions and substantially in equal amounts so that the primary conversion current is 0 in the period T1 to T2. .
[0012]
<Period T2-T3>
At time T2, the switching control circuit 70 simultaneously changes the second switching signal G2 and the third switching signal G3 from L to H, and the first low-side switch unit 101L and the second high-side switch unit 102H Turn on at the same time. On the other hand, both the first high-side switch unit 101H and the second low-side switch unit 102L are off. When the first low-side switch unit 101L and the second high-side switch unit 102H are turned on, the input voltage Vin is applied to the primary winding 3a of the transformer 3 in the reverse direction during the period T0 to T1. That is, the primary voltage Vt of the transformer 3 rapidly drops from 0 to −Vin. Therefore, the induced voltages of the secondary windings 3b and 3c of the transformer 3 rapidly drop from 0 to −Vin / n. At that time, the second rectifier diode 4c is already turned on. Therefore, the voltage Vb across the first rectifier diode 4b rapidly increases to +2 Vin / n when the direction of the arrow shown in FIG. 7, that is, the direction of the reverse bias is positive. Thereby, the first rectifier diode 4b is rapidly turned off. As a result, the secondary current of the transformer 3 starts to flow only through the second secondary winding 3c. Thus, commutation occurs on the secondary side of the transformer 3 at time T2.
[0013]
In the period T2 to T3, the first low-side switch unit 101L and the second high-side switch unit 102H are on. Thereby, the primary voltage Vt is substantially equal to −Vin. Further, the primary current It of the transformer 3 flows through the primary winding 3a from the second connection point Q to the first connection point P in the opposite direction to the periods T0 to T1. At that time, the voltage between both ends of each of the first secondary winding 3b and the second secondary winding 3c is substantially equal to -Vin / n. Since the first rectifier diode 4b is turned off and the second rectifier diode 4c is turned on, the voltage V5 across the smoothing inductor 5 is substantially equal to Vc−Vout = Vin / n−Vout. Therefore, the current I5 flowing through the smoothing inductor 5 increases linearly in the direction of the arrow shown in FIG. 6, as in the periods T0 to T1. Since the first rectifier diode 4b is off, the current I5 of the smoothing inductor 5 is substantially equal to the current Ic flowing through the second rectifier diode 4c. That is, in the period T2 to T3, the secondary current of the transformer 3 flows only through the second secondary winding 3c and increases linearly.
[0014]
As shown in FIG. 7, in the periods T2 to T3, the primary voltage Vt maintains a substantially constant value −Vin like the periods T0 to T1. Therefore, the excitation current included in the primary current It increases substantially linearly. However, the direction of the primary current It is opposite to the period T0 to T1. On the other hand, since the secondary current of the transformer 3 increases linearly as already described, the primary side converted current increases linearly. Accordingly, the primary current It increases substantially linearly.
[0015]
<Period T3-T4>
At time T3, the switching control circuit 70 simultaneously changes the second switching signal G2 and the third switching signal G3 from H to L, and simultaneously turns off the first low-side switch unit 101L and the second high-side switch unit 102H. To do. As a result, all four switch portions are turned off, so that the input voltage Vin is not applied to the primary winding 3a of the transformer 3. That is, the primary voltage Vt of the transformer 3 rapidly increases from the input voltage −Vin to 0. Therefore, the induced voltages of the secondary windings 3b and 3c of the transformer 3 are abruptly changed to zero. At that time, since the second rectifier diode 4c is already turned on, the voltage Vb across the first rectifier diode 4b suddenly changes to zero. Thereby, the first rectifier diode 4b is suddenly turned on. As a result, the secondary current of the transformer 3 starts to flow through both the first secondary winding 3b and the second secondary winding 3c, similarly to the time T1. Thus, commutation occurs at the secondary side of the transformer 3 at time T3.
[0016]
In the period T3 to T4, no voltage is applied to the primary winding 3a of the transformer 3, and the primary voltage Vt is zero. Therefore, the voltage V5 across the smoothing inductor 5 is substantially equal to the negative constant voltage −Vout. As a result, the current I5 of the smoothing inductor 5 decreases linearly with a gentle slope.
Since all the four switch units are turned off at time T3, the primary current It of the transformer 3 rapidly decreases to zero. Therefore, in the periods T3 to T4, as in the periods T1 to T2, the secondary current of the transformer 3 causes each of the two secondary windings 3b and 3c to be substantially opposite to each other so that the primary conversion current is 0. The same amount flows.
Thus, the state immediately before time T0 is reproduced in the periods T3 to T4. Thereafter, the operation in the period T0 to T4 described above is repeated.
[0017]
In hard switching by the switching control circuit 70, the voltage conversion rate, that is, the ratio between the input voltage Vin and the output voltage Vout is determined as follows:
Sum of the time of the first period T0 to T1 when the first high-side switch unit 101H is on and the time of the second period T2 to T3 when the second high-side switch unit 102H is on Is Ton. Since the voltage (Vin / n−Vout) is applied to the smoothing inductor 5 in the first period and the second period, the magnetic flux stored in the smoothing inductor 5 increases by (Vin / n−Vout) × Ton.
[0018]
On the other hand, the sum of the time of the periods T1 to T2 and the time of the periods T3 to T4 when all the switch units are off is Toff. Since the voltage (−Vout) is applied to the smoothing inductor 5 in each period, the magnetic flux stored in the smoothing inductor 5 decreases by Vout × Toff. Accordingly, the condition for balancing the increase and decrease of the magnetic flux of the smoothing inductor 5, that is, the reset condition of the smoothing inductor 5 is expressed by the following equation (1).
[0019]
(Vin / n−Vout) × Ton = Vout × Toff (1)
[0020]
From equation (1), the voltage conversion rate (ratio between input voltage Vin and output voltage Vout) is obtained by the following equation (2).
[0021]
Vout / Vin = δ / n where δ = Ton / (Ton + Toff) (2)
[0022]
As expressed by Equation (2), the output voltage Vout is stably maintained at a substantially constant value by controlling the ON / OFF time ratio δ with respect to the high-side switch unit.
[0023]
In recent years, demands for energy saving, size reduction, and weight reduction for various electronic devices have been remarkably increased. Accordingly, there is a strong demand for switching power supply devices that are highly efficient, downsized / lightweight, and improved in output stability. To meet such demands, the switching frequency must be further increased. However, the higher the switching frequency, the greater the switching loss. Therefore, in order to further increase the switching frequency, a switching technique capable of suppressing the switching loss is required. As such a switching technique, soft switching is known. Here, soft switching refers to the following switching: (1) During a transition from on to off or off to on, resonance occurs between the parasitic capacitor in the switch and the external inductor; (2 ) When either the resonant voltage or resonant current is zero, switch the switch from on to off or from off to on. In particular, switching performed when the voltage across the switch is zero is referred to as zero volt switching (ZVS).
[0024]
According to soft switching, power is not consumed by the switch when switching on / off. Therefore, in principle, there is no switching loss. In particular, according to ZVS, when the switch is turned on, no charge remains in the parasitic capacitor of the switch. Therefore, no surge current is generated.
[0025]
A conventional switching power supply device 110 that realizes ZVS is disclosed in, for example, Japanese Patent Laid-Open No. 11-89232. FIG. 8 shows the circuit configuration. The same components as those in the conventional example 100 shown in FIG. Differences from the conventional example 100 described above are the following two points: (1) The switch units 1H, 1L, 2H, and 2L in the full bridge are diodes 1HD, 1LD, and 2HD, respectively, together with the switch elements and parasitic capacitors. And 2LD. The diode is connected in parallel to the switch element. At that time, the cathode is connected to the high potential side, and the anode is connected to the low potential side. When the switch element is a transistor such as an IGBT, the diode may be the body diode of the transistor.
[0026]
(2) The switching control circuit 7 performs soft switching for the switch part in the full bridge. As a result, the switching loss is reduced as compared with the conventional example 100 described above. Hereinafter, soft switching by the switching control circuit 7 will be described.
FIG. 9 is a waveform diagram of current and voltage generated in each part of the circuit shown in FIG. 8 due to soft switching of the switching control circuit 7. In FIG. Here, the current and voltage of each circuit part are positive in the direction of the arrow shown in FIG.
The switching control circuit 7 outputs switching signals G1, G2, G3, and G4 to the switch elements 1HS, 1LS, 2HS, and 2LS, respectively. The switching signals G1, G2, G3, and G4 are each rectangular waves. The switch element is turned on while the switching signal shows a high potential (H), and the switch element is turned off while the switching signal shows a low potential (L).
[0027]
Soft switching by the switching control circuit 7 alternately realizes the following four periods at a predetermined time and in a predetermined cycle:
(1) The first period corresponds to periods T0 to T1 in FIG. In the first period, the first high-side switch unit 1H and the second low-side switch unit 2L are turned on, and the second high-side switch unit 2H and the first low-side switch unit 1L are turned off.
(2) The second period corresponds to periods T4 to T5 in FIG. In the second period, the first high-side switch unit 1H and the second low-side switch unit 2L are turned off, and the second high-side switch unit 2H and the first low-side switch unit 1L are turned on.
(3) The third period corresponds to periods T2 to T3 and periods T6 to T7 in FIG. 9, and is realized between the first period and the second period. In the third period, the first high-side switch unit 1H and the second high-side switch unit 2H are turned off, and the first low-side switch unit 1L and the second low-side switch unit 2L are turned on.
(4) The dead time is a minute period inserted between the above three periods, and corresponds to periods T1 to T2, T3 to T4, T5 to T6, and T7 to T8 in FIG. In the dead time, the pair of either the first high-side switch unit 1H and the first low-side switch unit 1L, or the second high-side switch unit 2H and the second low-side switch unit 2L is turned off. . That is, only one of the four switch parts is turned on, and the remaining three are turned off.
[0028]
Hereinafter, soft switching by the switching control circuit 7 will be described in the order of time from time T0 to time T8 shown in FIG. 9:
<Period T0-T1>
In the periods T0 to T1, the primary current It of the transformer 3 and the current I5 flowing through the smoothing inductor 5 both increase linearly as in the periods T0 to T1 in the conventional example described above. At that time, the secondary current of the transformer 3 flows only through the first secondary winding 3b.
[0029]
<Period T1-T2>
At time T1, the switching control circuit 7 changes the first switching signal G1 from H to L, and turns off the first high-side switch unit 1H. On the other hand, the second low-side switch unit 2L maintains the on state. At that time, resonance occurs in the leakage inductance of the primary winding 3a, the parasitic capacitor 1HC of the first high-side switch unit 1H, and the parasitic capacitor 1LC of the first low-side switch unit 1L. Due to the resonance, the primary current It charges the parasitic capacitor 1HC of the first high-side switch unit 1H and simultaneously discharges the parasitic capacitor 1LC of the first low-side switch unit 1L. Accordingly, the voltage V1H between both ends of the first high-side switch section 1H rises smoothly from 0, and the voltage V1L between both ends of the first low-side switch section 1L falls smoothly from the maximum value Vin.
[0030]
Immediately before time T2, the voltage V1H across the first high-side switch unit 1H reaches the maximum value Vin. At the same time, the voltage V1L across the first low-side switch unit 1L reaches zero. At that time, the diode 1LD of the first low-side switch unit 1L is turned on, and the voltage V1L between both ends is clamped to zero. At time T2, the switching control circuit 7 changes the second switching signal G2 from L to H, and turns on the first low-side switch unit 1L. In this way, ZVS is realized when the first low-side switch unit 1L is turned on.
[0031]
<Period T2-T3>
In the period T2 to T3, the primary winding 3a is short-circuited through the two low-side switch parts 1L and 2L. Meanwhile, the primary voltage Vt is substantially zero. Accordingly, commutation occurs on the secondary side of the transformer 3 as in the operation at the time T1 of the conventional example 100 described above. That is, since both the two rectifier diodes 4b and 4c are turned on, the secondary current flows through both the two secondary windings 3b and 3c. As a result, substantially only a constant voltage (−Vout) is applied to the smoothing inductor 5. Therefore, the current I5 flowing through the smoothing inductor 5, that is, the secondary current decreases linearly.
[0032]
Unlike the conventional example 100 described above, the secondary current flows through the first secondary winding 3b more than the second secondary winding 3c in the periods T2 to T3. The reason is as follows: Since the voltage Vc across the second rectifier diode 4c drops to 0 in the period T1 to T2, the second rectifier diode 4c is turned on at time T2. However, due to the leakage inductance on the secondary side of the transformer 3, the current Ib flowing through the first rectifier diode 4b decreases only slowly, and the current Ic flowing through the second rectifier diode 4c increases only slowly. Therefore, during the period T2 to T3, most of the secondary current flows through the first secondary winding 3b, and the remaining small portion flows through the second secondary winding 3c. As a result, since the cancellation between the secondary windings 3b and 3c is not sufficient, the primary-side converted current does not decrease so much in the period T2 to T3, and linearly decreases with a gentle slope due to the leakage inductance described above. . On the other hand, since the primary voltage Vt is substantially 0, the exciting current of the primary winding 3a is maintained substantially constant. As a result, the primary current It decreases linearly.
[0033]
<Period T3-T4>
At time T3, the switching control circuit 7 changes the fourth switching signal G4 from H to L, and turns off the second low-side switch unit 2L. On the other hand, the first low-side switch unit 1L maintains the on state. At that time, resonance occurs in the leakage inductance of the primary winding 3a, the parasitic capacitor 2HC of the second high-side switch unit 2H, and the parasitic capacitor 2LC of the second low-side switch unit 2L. Due to the resonance, the primary current It discharges the parasitic capacitor 2HC of the second high-side switch unit 2H. At the same time, the parasitic capacitor 2LC of the second low-side switch unit 2L is charged. Accordingly, the voltage V2H between both ends of the second high side switch section 2H falls smoothly from the maximum value Vin, and the voltage V2L between both ends of the second low side switch section 2L rises smoothly from zero.
[0034]
Immediately before time T4, the voltage V2L across the second low-side switch unit 2L reaches the maximum value Vin. At the same time, the voltage V2H across the second high-side switch portion 2H reaches 0. At that time, the diode 2HD of the second high side switch section 2H is turned on, and the voltage V2H between both ends is clamped to zero. At time T4, the switching control circuit 7 changes the third switching signal G3 from L to H, and turns on the second high-side switch unit 2H. Thus, ZVS is realized when the second high-side switch unit 2H is turned on.
[0035]
<Period T4-T5>
In the periods T4 to T5, the primary current It of the transformer 3 and the current I5 flowing through the smoothing inductor 5 both increase linearly as in the second periods T2 to T3 in the conventional example described above. However, the respective directions are opposite to those in the first period T0 to T1. Further, in the second period, the secondary current of the transformer 3 flows only through the second secondary winding 3c.
[0036]
<Period T5-T6>
At time T5, the switching control circuit 7 changes the third switching signal G3 from H to L, and turns off the second high-side switch unit 2H. On the other hand, the first low-side switch unit 1L maintains the on state. At that time, resonance occurs in the leakage inductance of the primary winding 3a, the parasitic capacitor 2HC of the second high-side switch unit 2H, and the parasitic capacitor 2LC of the second low-side switch unit 2L. Due to the resonance, the primary current It charges the parasitic capacitor 2HC of the second high-side switch unit 2H and simultaneously discharges the parasitic capacitor 2LC of the second low-side switch unit 2L. Therefore, the voltage V2H between both ends of the second high side switch section 2H rises smoothly from 0, and the voltage V2L between both ends of the second low side switch section 2L falls smoothly from the maximum value Vin.
[0037]
Immediately before time T6, the voltage V2H between both ends of the second high-side switch unit 2H reaches the maximum value Vin. At the same time, the voltage V2L across the second low-side switch unit 2L reaches zero. At that time, the diode 2LD of the second low-side switch unit 2L is turned on, and the voltage V2L between both ends is clamped to zero. At time T6, the switching control circuit 7 changes the fourth switching signal G4 from L to H, and turns on the second low-side switch unit 2L. In this way, ZVS is realized when the second low-side switch unit 2L is turned on.
[0038]
<Period T6-T7>
In the period T6 to T7, the primary winding 3a is short-circuited again through the two low-side switch portions 1L and 2L. Therefore, commutation occurs on the secondary side of the transformer 3 as in the operation in the periods T2 to T3, and the secondary current flows through both the two secondary windings 3b and 3c. However, the secondary current flows through the second secondary winding 3c more than the first secondary winding 3b, contrary to the periods T2 to T3. At that time, the primary conversion current does not decrease so much. Furthermore, substantially only a constant voltage (−Vout) is applied to the smoothing inductor 5. Accordingly, the primary current It and the secondary current I5 of the transformer 3 both decrease linearly.
[0039]
<Period T7 to T8>
At time T7, the switching control circuit 7 changes the second switching signal G2 from H to L, and turns off the first low-side switch unit 1L. On the other hand, the second low-side switch unit 2L maintains the on state. At that time, resonance occurs in the leakage inductance of the primary winding 3a, the parasitic capacitor 1HC of the first high-side switch unit 1H, and the parasitic capacitor 1LC of the first low-side switch unit 1L. Due to the resonance, the primary current It discharges the parasitic capacitor 1HC of the first high-side switch unit 1H. At the same time, the parasitic capacitor 1LC of the first low-side switch unit 1L is charged. Therefore, the voltage V1H between both ends of the first high-side switch section 1H smoothly falls from the maximum value Vin, and the voltage V1L between both ends of the first low-side switch section 1L increases smoothly from zero.
[0040]
Immediately before time T8, the voltage V1L across the first low-side switch unit 1L reaches the maximum value Vin. At the same time, the voltage V1H across the first high-side switch unit 1H reaches zero. At that time, the diode 1HD of the first high-side switch unit 1H is turned on, and the voltage V1H between both ends is clamped to zero. At time T8, the switching control circuit 7 changes the first switching signal G1 from L to H, and turns on the first high-side switch unit 1H. In this way, ZVS is realized when the first high-side switch unit 1H is turned on.
At time T8, the same state as time T0 is reproduced. Thus, the operation from time T0 to time T8 is repeated.
[0041]
The lengths of the periods T1 to T2, T3 to T4, T5 to T6, and T7 to T8, that is, the length of the dead time, are the first period T0 to T1, the second period T4 to T5, and , Sufficiently shorter than the respective times of the third period T2 to T3 and T6 to T7. In general, the length of the first period, the second period, and the third period is about several μsec, whereas the length of the dead time is about several tens to several hundreds nsec.
[0042]
If the dead time is ignored with respect to other periods such as the first period, the voltage conversion rate (ratio of the input voltage Vin to the output voltage Vout) in the soft switching is obtained as follows:
The sum of the time of the first period T0 to T1 and the time of the second period T4 to T5 is Ton. Since the voltage (Vin / n−Vout) is applied to the smoothing inductor 5 as described above in each period, the magnetic flux stored in the smoothing inductor 5 is combined in the above two periods (Vin / n−Vout). × Increases by Ton. On the other hand, the sum of the times of the third periods T1 to T4 and T5 to T8 is Toff. In each period, since the voltage (−Vout) is applied to the smoothing inductor 5 as described above, the magnetic flux stored in the smoothing inductor 5 is reduced by Vout × Toff. Therefore, the reset condition of the smoothing inductor 5 is expressed by the same expression (1) as that for hard switching.
[0043]
(Vin / n−Vout) × Ton = Vout × Toff (1)
[0044]
Therefore, the voltage conversion rate is expressed by the same equation (2) as in hard switching.
[0045]
Vout / Vin = δ / n where δ = Ton / (Ton + Toff) (2)
[0046]
That is, in the soft switching, similarly to the hard switching, the output voltage Vout is stably maintained at a substantially constant value by controlling the ON / OFF time ratio Δ between the high-side switch units 1H and 2H.
[0047]
In soft switching, the four switch sections 1H, 1L, 2H, and 2L are all turned on by ZVS as described above. Therefore, unlike hard switching, surge current / surge voltage associated with switching does not occur in the switch section. Thus, switching loss is reduced in soft switching compared to hard switching.
[0048]
[Problems to be solved by the invention]
In the conventional switching power supply device 100, the hard switching by the switching control circuit 70 has the following disadvantages: Each of the switch parts in the full bridge is connected in parallel with the switch element as shown in FIG. Including parasitic capacitors. When all the switch parts are off, the respective parasitic capacitors are stabilized in a state where charges are stored. In hard switching, one high side switch and one low side switch are turned on simultaneously. Therefore, when the switch unit is turned on, the parallel switch elements are turned on while the parasitic capacitor stores a certain amount of charge. At that time, the parasitic capacitor is short-circuited through the turned-on switch element and is rapidly discharged. As a result, a surge current is generated in the switch portion and converted into heat or electromagnetic waves. Electric power is dissipated outside through the heat and electromagnetic waves. Thus, power loss (switching loss) accompanying switching increases. Furthermore, the heat caused by the surge current fatigues the switch part, and the electromagnetic wave gives noise to surrounding circuit elements.
[0049]
For example, in FIG. 7, the voltage V1H and V2L between both ends of the first high-side switch unit 101H and the second low-side switch unit 102L are kept at finite values just before time T0. Accordingly, the parasitic capacitors 1HC and 2LC of the first high-side switch unit 101H and the second low-side switch unit 102L store charges proportional to the voltages V1H and V2L, respectively. When both switch parts are turned on at time T0, both parasitic capacitors 1HC and 2LC are suddenly discharged, and a surge current is generated. As a result, the current I1H flowing through the first high-side switch unit 101H and the current I2L flowing through the second low-side switch unit 102L increase so as to form a steep peak sc. When the first low-side switch unit 101L and the second high-side switch unit 102H are simultaneously turned on at time T2, similar peaks appear in the currents I1L and I2H flowing therethrough.
[0050]
In hard switching, the high-side switch unit and the low-side switch unit are simultaneously turned off one by one. Therefore, when the switch unit is turned off, the current supplied to the primary winding of the transformer is rapidly reduced. At that time, resonance occurs between the leakage inductance of the primary winding and the parasitic capacitor in the switch unit. Thereby, a surge voltage is generated. The generation of the surge voltage causes energy storage and dissipation due to leakage inductance of the primary winding, for example. Thus, switching loss increases.
[0051]
For example, in FIG. 7, substantially the same amount of current I1H = I2L flows immediately before the first high-side switch unit 101H and the second low-side switch unit 102L are turned off at time T1. When both switch sections are turned off at time T1, intense resonance occurs between the parasitic capacitors 1HC and 2LC and the leakage inductance of the primary winding 3a of the transformer 3, and a surge voltage is generated. Thereby, as shown in FIG. 7, the voltages V1H and V2L across the first high-side switch unit 101H and the second low-side switch unit 102L change so as to form a steep peak sv. . Similar peaks appear in both end-to-end voltages V1L and V2H when the first low-side switch unit 101L and the second high-side switch unit 102H are simultaneously turned on at time T3.
[0052]
In hard switching, as described above, a surge current / surge voltage is generated each time the switch unit is turned on / off, so that switching loss increases. An increase in switching loss is undesirable because it reduces the energy efficiency of the switching power supply. In addition, the surge current / surge voltage generates an electromagnetic wave of the switching frequency and its harmonics. Those electromagnetic waves are not preferable because they interfere with the operation of peripheral devices.
[0053]
Like the above-described switching power supply device 110, a conventional full-bridge converter using soft switching (hereinafter referred to as an active clamp full-bridge (ACFB) converter) is advantageous in terms of switching loss compared to the switching power supply device 100 using hard switching. . However, soft switching has the following disadvantages in terms of conduction loss compared to hard switching:
[0054]
Soft switching includes a third period, that is, a period in which the primary winding 3a of the transformer 3 is short-circuited (corresponding to periods T2 to T3 and T6 to T7 in FIG. 9). In the third period, the primary current It circulates through the two low-side switch parts 1L and 2L and the primary winding 3a. That is, in the third period, the primary current It is not 0 but finite. In that respect, soft switching is different from hard switching. The switch section generally includes an on-resistance, and the primary winding 3a of the transformer 3 generally includes a parasitic resistance. These resistors generate Joule heat when the primary current It is conducted, dissipating energy. Thus, in the soft switching, the conduction loss due to the primary current It in the third period (hereinafter referred to as the circulating current loss) increases compared to the hard switching.
[0055]
Due to the demand for miniaturization of the switching power supply device, the size of the switch unit is limited. As a result, there is a lower limit to the on-resistance of the switch unit. The lower limit is larger as the switching power supply device is smaller. In such a small switching power supply device, the increase in circulating current loss can exceed the reduction in switching loss due to soft switching. At that time, the efficiency of the switching power supply is not sufficiently improved.
[0056]
In ACFB converter 110, the leakage inductance of the primary winding of the transformer is used for resonance with the parasitic capacitor of the switch unit. However, such leakage inductance is inherently small. When the leakage inductance is too small, a sufficiently large resonance current cannot be obtained, and the discharge and charging of the parasitic capacitor of the switch unit can be insufficient. As a result, ZVS cannot be realized and switching loss may increase.
[0057]
The source of surge current / surge voltage that accompanies switching is not limited within the switch section. In both of the conventional switching power supply device 100 by hard switching and the ACFB converter 110 by soft switching, surge current / surge voltage is generated when the rectifier diodes 4b and 4c are turned off at the time of commutation on the secondary side. . Accordingly, as shown in FIGS. 7 and 9, respectively, a peak cf is generated in the currents Ib and Ic flowing through the rectifier diodes 4b and 4c, and a peak vf is generated in the voltages Vb and Vc between both ends. The causes of surge current / surge voltage generation in rectifier diodes 4b and 4c are as follows:
[0058]
At the start of each of the first period and the second period, one of the two rectifier diodes 4b and 4c is turned off. The start times of the first period and the second period correspond to times T0 and T2 in FIG. 7, and correspond to times T0 and T4 in FIG. 9, respectively. The diode generally accumulates charge while it is on. As a result, the accumulated charge is discharged as a reverse current when turned off by applying a reverse bias. In the switching power supply devices 100 and 110 described above, when any of the rectifier diodes 4b and 4c is turned off by the commutation on the secondary side, a surge current is generated as a reverse current.
[0059]
The two rectifier diodes 4b and 4c are connected to the two secondary windings 3b and 3c of the transformer 3, respectively. For example, when the first rectifier diode 4b is turned off, the parasitic capacitance of the first rectifier diode 4b and the leakage inductance of the first secondary winding 3b resonate based on the reverse current. Thereby, a surge voltage is generated between both ends of the first rectifier diode 4b. Similarly, when the second rectifier diode 4c is turned off, a surge voltage based on a reverse current is generated between both ends thereof.
The surge current / surge voltage accompanying the on / off of the rectifier diode generates noise. The power loss due to the noise is undesirable because it reduces the efficiency of the switching power supply as part of the switching loss.
[0060]
An object of this invention is to provide the switching power supply device which can reduce a switching loss by soft switching and can reduce a circulating current loss.
[0061]
[Means for Solving the Problems]
The switching power supply device according to the present invention includes:
(A) (a) Four switch parts, all of which are (1) a switch element that is turned on and off by an external switching signal, (2) a diode connected in parallel to the switch element, and (3 A first high-side switch unit, a second high-side switch unit, a first low-side switch unit, and a second low-side switch unit, each including a capacitor connected in parallel to the switch element;
(b) (1) a transformer comprising: a primary winding; and (2) a first secondary winding and a second secondary winding that are connected in series with each other at a common end thereof;
a first rectifying unit and a second rectifying unit, each including: (c) (1) a rectifying element; and (2) a snubber connected in parallel to the rectifying element and including a snubber capacitor;
(d) a smoothing unit including a first input terminal and a second input terminal, and smoothing and outputting the input through the input terminals;
A DC-DC converter having:
(B) The cathode of the first high side switch section is connected to the high potential terminal of the substantial DC constant voltage source, the anode of the first high side switch section is connected to the cathode of the first low side switch section, and the first low side switch The anode of the switch part is connected to the low potential terminal of the substantial DC constant voltage source, respectively;
(C) The cathode of the second high-side switch section is connected to the high potential terminal of the substantial DC constant voltage source, the anode of the second high-side switch section is connected to the cathode of the second low-side switch section, and the second low-side switch section The anode of the switch part is connected to the low potential terminal of the substantial DC constant voltage source, respectively;
(D) One end of the primary winding of the transformer is connected to the connection point between the first high-side switch unit and the first low-side switch unit, and the other end is connected to the second high-side switch unit and the second low-side switch unit. Connect to each connection point;
(E) An end different from the common end of the first secondary winding of the transformer is connected to one end of the first rectification unit, and an end different from the common end of the second secondary winding is the first rectification. A common end connected to one end of a second rectifying unit having the same polarity as one end of the unit, and a first input terminal of the smoothing unit;
(F) The other ends of the first rectification unit and the second rectification unit are connected to the second input terminal of the smoothing unit;
DC-DC converter;
And
(A) Delay time based on the resonance period determined by the equivalent capacity of the snubber of each of the first rectifying unit and the second rectifying unit and the leakage inductance on the secondary side of the transformer when the primary winding is short-circuited Determine;
(B) A switching signal is output to each of the four switch sections at a predetermined switching frequency and phase, thereby
(a) (1) a first period of turning on the first high-side switch unit and the second low-side switch unit and turning off the second high-side switch unit and the first low-side switch unit; And (2) a second period in which the first high-side switch unit and the second low-side switch unit are turned off and the second high-side switch unit and the first low-side switch unit are turned on. Each realized alternately at a predetermined time and a predetermined period;
(b) At the end of the first period, one of the first high-side switch unit and the second low-side switch unit is turned off with a delay time after the other off;
(c) At the end of the second period, one of the second high-side switch unit and the first low-side switch unit is turned off with a delay time after the other off;
Switching controller for;
It comprises.
[0062]
In the switching power supply device described above, both the high-side switch unit and the low-side switch unit that were turned on are turned off when the delay time has elapsed since the end of each of the first period and the second period. At that time, the current (primary current) flowing through the primary winding of the transformer is quickly attenuated as follows.
[0063]
One end of the first rectifier connected to the first secondary winding and one end of the second rectifier connected to the second secondary winding have the same polarity. That is, when the anode of the first rectifying unit is connected to the first secondary winding, the anode of the second rectifying unit is connected to the second secondary winding. Conversely, when the cathode of the first rectifying unit is connected to the first secondary winding, the cathode of the second rectifying unit is connected to the second secondary winding.
[0064]
The first rectification unit is turned off in the first period, and the second rectification unit is turned off in the second period. Thereby, full-wave rectification is realized on the secondary side of the transformer. At that time, the internal snubber capacitor accumulates electric charge by reverse bias in one of the rectifiers turned off. At the end of each of the first period and the second period, the respective induced voltage of the secondary winding is reduced to zero. At that time, the snubber capacitor is discharged in one of the rectifiers that has been turned off. Due to the discharge current, current starts to flow through the secondary winding connected to the rectifier before the rectifier in the rectifier is turned on. Based on the current, the snubber capacitor and the leakage inductance on the secondary side of the transformer resonate. Here, the leakage inductance on the secondary side of the transformer refers to a leakage inductance that is equivalently generated between each end different from the shared end of the two secondary windings when the primary winding is short-circuited. In the present specification, the leakage inductance is referred to as a leakage inductance on the secondary side of the transformer when the primary winding is short-circuited. Therefore, unlike the conventional switching power supply device, the commutation of the secondary current of the transformer proceeds smoothly and quickly. That is, immediately after the end of the first period and the second period, the secondary current begins to flow in substantially equal amounts in both secondary windings. As a result, the primary side converted current is canceled out smoothly and quickly between the secondary currents respectively flowing through the two secondary windings. Accordingly, at the end of each of the first period and the second period, the primary current is smoothly and rapidly reduced. Therefore, when the primary current is cut off with the delay time, the surge voltage is reduced, so that the switching loss due to the cutoff of the primary current is reduced. Furthermore, since the primary current is interrupted between the first period and the second period, the circulating current loss due to the primary current is reduced.
[0065]
In the above switching power supply device, the switching controller is
(A) At the end of the first period, the second low-side switch part is turned off with a delay after the first high-side switch part is turned off;
(B) At the end of the second period, the first low-side switch unit may be turned off with a delay after the second high-side switch unit is turned off.
In this switching power supply device, the primary current of the transformer circulates through the two low-side switch sections and the primary winding from the end of each of the first period and the second period until the lapse of the corresponding delay time.
[0066]
In addition, the switching controller
(A) At the end of the first period, the first high-side switch part is turned off with a delay time after turning off the second low-side switch part;
(B) At the end of the second period, the second high-side switch unit may be turned off with a delay after the first low-side switch unit is turned off.
In this switching power supply device, the primary current of the transformer circulates between the two high-side switch units and the primary winding from the end of each of the first period and the second period to the lapse of the corresponding delay time. .
[0067]
In addition to the above, the switching control unit
(A) At the end of the first period, the second low-side switch part is turned off with a delay after the first high-side switch part is turned off;
(B) At the end of the second period, the second high-side switch unit may be turned off with a delay after the first low-side switch unit is turned off.
In this switching power supply device, the primary current of the transformer circulates between the two low-side switch units and the primary winding from the end of the first period to the lapse of the delay time. On the other hand, the primary current of the transformer circulates between the two high-side switch units and the primary winding from the end of the second period until the lapse of the delay time. Since the circulating parts are alternately switched in this way, the stress applied to each of the switch parts is equalized.
[0068]
In the above switching power supply device, the switching controller is
(A) The first dead time is determined based on the resonance period determined by the equivalent capacity of the first high-side switch section, the equivalent capacity of the first low-side switch section, and the leakage inductance of the primary winding of the transformer. ;
(B) The second dead time is determined based on the resonance period determined by the equivalent capacity of the second high-side switch part, the equivalent capacity of the second low-side switch part, and the leakage inductance of the primary winding of the transformer. ;
(C) at the end of each of the first period and the second period,
(a) turn off both the first high-side switch and the first low-side switch for only the first dead time; or
(b) The second high-side switch unit and the second low-side switch unit may be turned off together for the second dead time.
[0069]
  For example, at the end of the first period, the first high-side switch unit is turned off before the second low-side switch unit. At that time, resonance occurs between the leakage inductance of the primary winding of the transformer and the equivalent capacitances of the first high-side switch unit and the first low-side switch unit. Due to the resonance, the capacitor in the first high-side switch unit is charged, and the capacitor in the first low-side switch unit is discharged. Furthermore, when the capacitor in the first low-side switch unit completes the discharge, the diode in the same switch unit is turned on, and the voltage across both ends is clamped to zero. In that stateFirstSince the low-side switch can be turned on,FirstZVS can be realized when the low-side switch is turned on. Therefore, the switching loss can be reduced. When the second low-side switch unit is turned off first at the end of the first period, and at the end of the second period, ZVS can be similarly realized for the switch unit that is turned on first.
[0070]
The dead time is determined based on the resonance period determined by the equivalent capacitance of the high-side switch unit, the equivalent capacitance of the low-side switch unit, and the leakage inductance of the primary winding of the transformer. Preferably, it is substantially equal to 1/4 of its resonance period. On the other hand, the delay time is determined based on a resonance period determined by the equivalent capacity of the snubber and the leakage inductance on the secondary side of the transformer when the primary winding is short-circuited. Preferably it is substantially equal to 1/4 of its resonance period. The delay time is mainly adjusted by the capacity of the snubber capacitor. Therefore, in the above switching power supply device, the primary current can be cut off immediately after the above ZVS is realized. Thus, the circulating current loss as well as the switching loss can be reduced.
[0071]
In the above switching power supply device, the snubber may include a resistor connected in series to the snubber capacitor. At the start of each of the first period and the second period, an induced voltage from the primary side is generated in each secondary winding of the transformer. As a result, one of the two rectifying units that have been turned on is turned off. At that time, in the rectifying unit that is turned off, resonance occurs between the snubber capacitor and the leakage inductance on the secondary side of the transformer. The resistance in the snubber quickly attenuates the resonance. Thus, the occurrence of linking due to resonance is suppressed, and the peak of the voltage applied to the rectifier element parallel to the snubber is reduced.
[0072]
The switching power supply further includes (a) a sub-rectifier element in which the snubber is connected in parallel to the resistor; (b) one of the anodes of the rectifier element and the sub-rectifier element and the cathodes are connected. You may do it. At the end of each of the first period and the second period, the snubber capacitor is discharged in the rectifying unit that has been turned off. The discharge current at that time flows through a sub-rectifier element parallel to the resistor. Thereby, reduction of the discharge current due to the resistance is avoided. Therefore, the peak of the resonance current between the equivalent capacity of the snubber and the leakage inductance on the secondary side of the transformer increases. As a result, at the end of each of the first period and the second period, the primary side converted current is sufficiently canceled more quickly, and the primary current is sufficiently reduced. Therefore, not only the conduction loss due to the resistance in the snubber but also the switching loss due to the primary current and the circulating current loss can be further reduced. In addition, since the heat generation of the resistance in the snubber due to the discharge current is suppressed, it is possible to prevent an increase in resistance value due to heat and a deterioration in the functions of other elements.
[0073]
In the above switching power supply device, the rectifying element and the sub-rectifying element included in the snubber are preferably diodes. In addition, semiconductor switch elements such as IGBTs and MOSFETs may be used. At that time, switching of these semiconductor switch elements is preferably controlled by the switching control unit in synchronization with the switch unit in the primary-side full bridge.
[0074]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings, taking preferred examples.
[0075]
Example 1
The switching power supply device 10 according to the first embodiment of the present invention includes a full bridge converter. The switching control circuit 7 performs hard switching to turn on the four switch units 1H, 1L, 2H, and 2L constituting the full bridge.
FIG. 1 shows a circuit of a switching power supply device 10 according to the first embodiment. As apparent from the comparison between FIG. 1 and FIG. 8, the circuit configuration of the switching power supply 10 according to the first embodiment is a conventional switching power supply in that it has two snubber capacitors 8Cb and 8Cc on the secondary side of the transformer 3. For example, it is different from the conventional example 110 shown in FIG.
[0076]
A DC power source is connected to the input terminals 1a and 1b. Here, the DC power supply may be a substantial one obtained by rectifying the AC from the AC power supply. The input voltage from the DC power supply is Vin. Further, the high potential side of the DC power source is connected to the first input terminal 1a, and the low potential side is connected to the second input terminal 1b.
[0077]
Each of the four switch units 1H, 1L, 2H, and 2L includes a switch element, a diode, and a capacitor connected in parallel between the two terminals. In particular, due to the polarity of the diode, the switch unit has the same polarity as the diode. The terminals of the switch section connected to the cathode and anode of the diode are called the cathode and anode, respectively. Furthermore, the characteristics of each circuit element are common to the four switch sections.
[0078]
The four switch parts are connected in series two by two to form two pairs. Furthermore, these pairs are connected in parallel. Of the two switch parts included in each pair, one connected to the first input terminal 1a is called the high side, and the other connected to the second input terminal 1b is called the low side. In FIG. 1, a first high-side switch unit 1H and a first low-side switch unit 1L, and a second high-side switch unit 2H and a second low-side switch unit 2L are connected in series. The anode of the first high-side switch unit 1H and the cathode of the first low-side switch unit 1L are the first connection point P, the anode of the second high-side switch unit 2H and the cathode of the second low-side switch unit 2L Are connected at the second connection point Q. The cathode of the first high-side switch unit 1H and the cathode of the second high-side switch unit 2H are both connected to the first input terminal 1a. The anode of the first low side switch unit 1L and the anode of the second low side switch unit 2L are both connected to the second input terminal 1b.
[0079]
The switch elements 1HS, 1LS, 2HS, and 2LS are preferably MOSFETs. In addition, a bipolar transistor or an IGBT may be used.
The diodes 1HD, 1LD, 2HD, and 2LD are preferably parasitic diodes of the switch elements 1HS, 1LS, 2HS, and 2LS in the same switch section. In addition, the switch elements 1HS, 1LS, 2HS, and 2LS may be independent elements.
Four capacitors 1HC, 1LC, 2HC, and 2LC are parasitic capacitors of the switch elements 1HS, 1LS, 2HS, and 2LS in the same switch unit, and can be regarded as being connected in parallel to the corresponding switch elements. Capacitors 1HC, 1LC, 2HC, and 2LC equivalently include a wiring and a parasitic capacitor of transformer 3. Furthermore, the capacitors 1HC, 1LC, 2HC, and 2LC may include a capacitor as an independent element different from the switch element.
[0080]
The primary winding 3a of the transformer 3 is connected between the connection points P and Q of each pair of switch units. The transformer 3 has two secondary windings 3b and 3c. Here, the turns ratio of the primary winding 3a, the first secondary winding 3b, and the second secondary winding 3c of the transformer 3 is n: 1: 1 (n is a positive real number). The common end R, which is a connection point between the secondary windings 3b and 3c, is connected to the second output terminal 2b. In each of the first secondary winding 3b and the second secondary winding 3c, the end opposite to the shared end R is connected to the respective anodes of the first rectifier diode 4b and the second rectifier diode 4c. The The cathodes of the first rectifier diode 4b and the second rectifier diode 4c are both connected to one end of the smoothing inductor 5. The two rectifier diodes 4b and 4c perform full-wave rectification on the output of the transformer 3.
[0081]
The other end of the smoothing inductor 5 is connected to the first output terminal 2a. The inductance of the smoothing inductor 5 is sufficiently larger than other circuit elements such as the transformer 3. In the smoothing capacitor 6, one end is connected between the first output terminal 2a and the smoothing inductor 5, and the other end is connected to the second output terminal 2b. The capacity of the smoothing capacitor 6 is sufficiently larger than the capacitors in the switch section and the snubber capacitors 8Cb and 8Cc. The smoothing inductor 5 and the smoothing capacitor 6 constitute a smoothing filter, and smooth the output of the transformer 3 rectified through the two rectifier diodes 4b and 4c. Thus, the voltage across the smoothing capacitor 6, that is, the output voltage Vout between the two output terminals 2a and 2b is substantially constant.
[0082]
The first snubber capacitor 8Cb and the second snubber capacitor 8Cc are connected in parallel to the first rectifier diode 4b and the second rectifier diode 4c, respectively. Thus, in the first embodiment, the first rectifier diode 4b and the first snubber capacitor 8Cb serve as the first rectifier unit, and the second rectifier diode 4c and the second snubber capacitor 8Cc serve as the second rectifier unit. Configure each. Here, the capacitances of the snubber capacitors 8Cb and 8Cc are substantially equal to each other, and are sufficiently larger than the capacitances of the parasitic capacitors 1HC, 1LC, 2HC, and 2LC in the switch section.
[0083]
The switching control circuit 7 is connected to the two output terminals 2a and 2b, respectively, and measures the output voltage Vout. Further, the switching control circuit 7 is connected to the switch elements 1HS, 1LS, 2HS, and 2LS, and controls on / off of the respective switch elements by switching signals G1, G2, G3, and G4. Each switch element is preferably a MOSFET. At that time, the switching control circuit 7 is connected to the gates of the respective switching elements, and outputs voltage signals to the respective gates as the switching signals G1 to G4. Thereby, each gate potential is changed to switch on / off of each switch element. In particular, the switching control circuit 7 adjusts the frequency (switching frequency) of the switching signals G1 to G4 based on the measured output voltage Vout. Thereby, the ON / OFF time ratio of the switch elements 1HS, 1LS, 2HS, and 2LS is controlled, and the output voltage Vout is kept constant.
[0084]
The switching power supply device 10 according to the first embodiment performs the following switching on the four switch units having a full bridge configuration:
In the switching power supply apparatus 10, as shown in FIG. 1, the input voltage Vin is applied between the two input terminals 1a and 1b, and the output voltage Vout is applied between the two output terminals 2a and 2b. The input voltage Vin and the output voltage Vout are each maintained substantially constant.
[0085]
FIG. 2 is a waveform diagram of current and voltage generated in each part of the circuit shown in FIG.
The switching signals G1, G2, G3, and G4 are rectangular waves, each having a predetermined width. While the switching signals G1, G2, G3, and G4 indicate a high potential (H), the switch elements 1HS, 1LS, 2HS, and 2LS are turned on. On the contrary, while the switching signals G1, G2, G3, and G4 indicate the low potential (L), the switch elements 1HS, 1LS, 2HS, and 2LS are turned off.
[0086]
The voltages V1H, V1L, V2H, and V2L are voltages applied to the switch units 1H, 1L, 2H, and 2L, respectively. Each polarity is defined as positive in the direction of the arrow shown in FIG. 1, that is, the potential of the end (cathode) on the first input terminal 1a side with respect to the end (anode) on the second input terminal 1b side. The The voltage VH across the high side switch is substantially equal to the difference between the input voltage Vin and the voltage VL across the same pair of low side switches: VH = Vin−VL. Accordingly, FIG. 2 shows only the voltages V1L and V2L across the low-side switch units 1L and 2L.
[0087]
Currents I1H, I1L, I2H, and I2L are currents that flow through the switch units 1H, 1L, 2H, and 2L, respectively. Each polarity is defined as positive in the direction of the arrow shown in FIG. 1, ie, from the cathode to the anode.
[0088]
The primary voltage Vt is a voltage applied to the primary winding 3a of the transformer 3. The polarity is defined as positive in the direction of the arrow shown in FIG. 1, that is, the potential of the first connection point P with respect to the second connection point Q. The primary voltage Vt is substantially equal to the difference between the voltages V1L and V2L across the two low-side switch sections 1L and 2L, respectively: Vt = V1L−V2L.
[0089]
The primary current It is a current flowing through the primary winding 3a of the transformer 3. The polarity is defined as positive in the direction of the arrow shown in FIG. 1, that is, the direction from the first connection point P to the second connection point Q. The primary current It is substantially equal to the difference between the current flowing through the high-side switch section and the current flowing through the low-side switch section: It = I1H−I1L = −I2H + I2L.
[0090]
The voltages Vb and Vc are voltages across the first rectifier diode 4b and the second rectifier diode 4c, respectively. Each polarity is defined with the direction of the arrow shown in FIG. 1 as positive. That is, the reverse bias direction is positive for each rectifier diode. Since the turns ratio of the primary winding 3a and the secondary windings 3b and 3c is n: 1: 1, the voltage across the two secondary windings 3b and 3c is 1 / n times the primary voltage Vt = Substantially equal to Vt / n.
[0091]
The secondary currents Ib and Ic are currents flowing through the two secondary windings 3b and 3c of the transformer 3, respectively. Each polarity is defined with the direction of the arrow shown in FIG. 1 as positive. That is, the first secondary current Ib is directed from the common end R of the first secondary winding 3b to the other end, and the second secondary current Ic is from the other end of the second secondary winding 3c. The direction toward the shared end R is positive.
The current I5 is a current flowing through the smoothing inductor 5, and is substantially equal to the output current of the switching power supply device 10. The polarity is defined as positive in the direction of the arrow shown in FIG. 1, that is, the direction from the end on the transformer 3 side to the end on the first output terminal 2a side. The output current I5 is substantially equal to the sum of the two secondary currents Ib and Ic: I5 = Ib + Ic.
[0092]
The primary current It is equal to the sum of the exciting current for the transformer 3 and the primary side converted current determined by the secondary current of the transformer 3. The primary side conversion current is obtained from the two secondary currents Ib and Ic of the transformer 3 as follows: (1) Regarding the mutual inductance between the primary winding 3a and the first secondary winding 3b, Find the ratio to the self-inductance; (2) Find the product of the ratio and the first secondary current Ib; (3) Find the product in the same way for the second secondary winding 3c; (4) First The sum of the product obtained for the secondary winding 3b and the product obtained for the second secondary winding 3c is obtained. The sum is substantially equal to the primary conversion current. The primary-side converted current in Example 1 is substantially equal to 1 / n times the difference between the first secondary current Ib and the second secondary current Ic = (Ib−Ic) / n.
[0093]
The switching control circuit 7 performs hard switching when the four switch units are turned on. That is, the high-side switch unit and the low-side switch unit are simultaneously turned on one by one. On the other hand, as will be described later, one switch off is delayed by a delay time from the other switch switch off. Switching by the switching control circuit 7 alternately realizes the following four periods at a predetermined time and in a predetermined cycle:
[0094]
(1) The first period corresponds to periods T0 to T1 in FIG. In the first period, the first high-side switch unit 1H and the second low-side switch unit 2L are turned on, and the second high-side switch unit 2H and the first low-side switch unit 1L are turned off.
(2) The second period corresponds to periods T4 to T5 in FIG. In the second period, the first high-side switch unit 1H and the second low-side switch unit 2L are turned off, and the second high-side switch unit 2H and the first low-side switch unit 1L are turned on.
(3) The third period corresponds to the periods T3 to T4 and the periods T7 to T8 in FIG. 2, and is realized between the first period and the second period. In the third period, all four switch parts are turned off.
(4) The fourth period corresponds to periods T1 to T3 and periods T5 to T7 in FIG. 2, and is realized at the end of each of the first period and the second period. At the start of the fourth period, only one of the two switch sections that were turned on immediately before is turned off first. At the end of the fourth period, the other switch unit is turned off. The length of the fourth period is called the delay time.
[0095]
Hereinafter, switching by the switching control circuit 7 will be described in the order of time from time T0 to time T8 shown in FIG.
<Period T0-T1>
At time T0, the switching control circuit 7 simultaneously changes the first switching signal G1 and the fourth switching signal G4 from L to H, and simultaneously switches the first high-side switch unit 1H and the second low-side switch unit 2L. Turn on. On the other hand, both the first low-side switch unit 1L and the second high-side switch unit 2H are off.
[0096]
During the period T0 to T1, a substantially constant and positive input voltage Vin is applied to the primary winding 3a of the transformer 3 through the first high-side switch unit 1H and the second low-side switch unit 2L. That is, the voltage Vt across the primary winding 3a is substantially equal to the input voltage Vin. Further, the primary current It of the transformer 3 flows through the primary winding 3a from the first connection point P to the second connection point Q, that is, in the direction of the arrow shown in FIG. At that time, a positive voltage Vin / n is induced in the secondary windings 3b and 3c of the transformer 3, respectively. At that time, since the first rectifier diode 4b is on, the voltage V5 across the smoothing inductor 5 is substantially equal to Vin / n-Vout. Therefore, the output current I5 flowing through the smoothing inductor 5 increases linearly in the direction of the arrow shown in FIG. 1 during the period T0 to T1. However, since the inductance of the smoothing inductor 5 is sufficiently large, the increase in the current I5 flowing through the smoothing inductor 5 is moderate. Since the voltage Vc across the second rectifier diode 4c is substantially equal to + 2Vin / n and positive, the second rectifier diode 4c is off. Therefore, the current I5 of the smoothing inductor 5 is substantially equal to the first secondary current Ib. That is, in the period T0 to T1, the secondary current of the transformer 3 flows only through the first secondary winding 3b and increases linearly. When the second rectifier diode 4c is off, the second snubber capacitor 8Cc stores a charge proportional to the reverse bias Vc = 2Vin / n with respect to the second rectifier diode 4c.
[0097]
As described above, the primary current It of the transformer 3 is substantially equal to the sum of the exciting current and the primary side converted current. As shown in FIG. 2, the primary voltage Vt maintains a substantially constant value Vin in the periods T0 to T1. Therefore, the excitation current included in the primary current It increases substantially linearly. On the other hand, since the secondary current of the transformer 3 increases linearly as already described, the primary side converted current increases linearly. Accordingly, the primary current It increases substantially linearly.
[0098]
<Period T1-T3>
At time T1, the switching control circuit 7 changes the first switching signal G1 from H to L, and turns off the first high-side switch unit 1H. On the other hand, the second low-side switch unit 2L maintains the on state. At that time, resonance occurs in the leakage inductance of the primary winding 3a, the parasitic capacitor 1HC of the first high-side switch unit 1H, and the parasitic capacitor 1LC of the first low-side switch unit 1L. Due to the resonance, the primary current It charges the parasitic capacitor 1HC of the first high-side switch unit 1H and simultaneously discharges the parasitic capacitor 1LC of the first low-side switch unit 1L. Accordingly, the voltage V1H between both ends of the first high-side switch section 1H rises smoothly from 0, and the voltage V1L between both ends of the first low-side switch section 1L falls smoothly from the maximum value Vin.
[0099]
At time T2, the voltage V1H across the first high-side switch unit 1H reaches the maximum value Vin. At the same time, the voltage V1L across the first low-side switch unit 1L reaches zero. At that time, the diode 1LD of the first low-side switch unit 1L is turned on, and the voltage V1L between both ends is clamped to zero. As a result, the primary voltage Vt similarly falls smoothly and is clamped at zero.
[0100]
In each of the first secondary winding 3b and the second secondary winding 3c, the voltage between both ends falls with the fall of the primary voltage Vt in the period T1 to T2. At that time, since the first rectifier diode 4b is on, the voltage Vb between both ends is substantially zero. On the other hand, the voltage Vc across the second rectifier diode 4c drops. That is, the reverse bias with respect to the second rectifier diode 4c decreases. Thereby, the second snubber capacitor 8Cc is discharged. Based on the discharge current, the second snubber capacitor 8Cc and the leakage inductance on the secondary side of the transformer 3 start to resonate. Here, the leakage inductance on the secondary side is the secondary side of the transformer 3 when the primary winding 3a is short-circuited, the end different from the common end R of the first secondary winding 3b, and the second secondary inductance. This is substantially equal to the leakage inductance that occurs equivalently between the shared winding end R and another end of the next winding 3c. Hereinafter, this leakage inductance is referred to as a leakage inductance on the secondary side of the transformer when the primary winding is short-circuited. Furthermore, the capacitance of the second snubber capacitor 8Cc is sufficiently larger than the capacitance of the parasitic capacitor in the switch section. Therefore, the resonance period between the second snubber capacitor 8Cc and the above-described secondary side leakage inductance is sufficiently longer than the resonance period between the parasitic capacitor in the switch section and the primary winding 3a. As a result, the voltage Vc across the second rectifier diode 4c gradually falls below the primary voltage Vt after time T1, and reaches 0 immediately before time T3 after time T2. At that time, the second rectifier diode 4c is turned on.
[0101]
The second snubber capacitor 8Cc is discharged by resonance in the periods T1 to T3. The discharge current rapidly and smoothly increases the second secondary current Ic flowing through the second secondary winding 3c. As a result, the first secondary current Ib flowing through the first secondary winding 3b decreases rapidly and smoothly. As a result, there is almost no difference between the secondary currents Ib and Ic flowing through the secondary windings 3b and 3c, respectively, near time T3. In this way, in the period T1 to T3, the secondary currents of the secondary windings 3b and 3c are quickly canceled out, so that the primary side converted current is rapidly and smoothly reduced to substantially zero. Therefore, the primary current It decreases rapidly and smoothly from the value at time T1.
[0102]
<Period T3-T4>
At time T3, the switching control circuit 7 changes the fourth switching signal G4 from H to L, and turns off the second low-side switch unit 2L. As a result, all four switch sections are turned off. At that time, the voltages V1L and V2L across the two low-side switch sections 1L and 2L are both zero. Accordingly, the voltages V1H and V2H across the two high-side switch units 1H and 2H are equal to the input voltage Vin. After time T3, the voltage across the high-side switch section drops from Vin with ringing, and the voltage across the low-side switch section rises from 0 with ringing.
[0103]
The primary voltage Vt of the transformer 3 is substantially maintained at 0 in the periods T3 to T4. Accordingly, the induced voltages of the two secondary windings 3b and 3c of the transformer 3 are also maintained at 0. Further, both the first rectifier diode 4b and the second rectifier diode 4c are on. Therefore, the voltage V5 across the smoothing inductor 5 is maintained at a negative constant voltage −Vout. As a result, the output current I5 flowing through the smoothing inductor 5 decreases smoothly and linearly with a gentle slope.
[0104]
The primary current It of the transformer 3 once decreases rapidly in the periods T1 to T3, and then decreases to 0 more rapidly at time T3. Therefore, since the decrease is slow compared with the conventional hard switching, noise such as undershoot is reduced.
After time T3, the primary current It does not substantially flow. Therefore, the secondary current of the transformer 3 continues to flow through the two secondary windings 3b and 3c in opposite directions and substantially in equal amounts so that the primary side converted current is zero. That is, in the periods T3 to T4, the first secondary current Ib and the second secondary current Ic are substantially equal.
[0105]
<Period T4-T5>
At time T4, the switching control circuit 7 simultaneously changes the second switching signal G2 and the third switching signal G3 from L to H, and simultaneously switches the second low-side switch unit 1L and the second high-side switch unit 2H. Turn on. On the other hand, the first high-side switch unit 1H and the second low-side switch unit 2L are both turned off. When the first low-side switch unit 1L and the second high-side switch unit 2H are turned on, the input voltage Vin is applied to the primary winding 3a of the transformer 3 in the reverse direction during the period T0 to T1, and the primary voltage Vt is applied. Descent rapidly from 0 to -Vin. Further, the primary current It of the transformer 3 flows through the primary winding 3a from the second connection point Q to the first connection point P in the opposite direction to the periods T0 to T1. Therefore, the respective induced voltages of the secondary windings 3b and 3c of the transformer 3 rapidly drop from 0 to −Vin / n. At that time, since the second rectifier diode 4c is already turned on, the voltage Vb across the first rectifier diode 4b rapidly rises to + 2Vin / n. Thereby, the first rectifier diode 4b is turned off. As a result, the secondary current of the transformer 3 starts to flow only through the second secondary winding 3c. Thus, commutation occurs on the secondary side of the transformer 3 at time T4.
[0106]
Further, in the periods T4 to T5, the voltage V5 across the smoothing inductor 5 is substantially equal to Vin / n−Vout, as in the periods T0 to T1. Accordingly, the output current I5 flowing through the smoothing inductor 5 increases linearly and gently in the direction of the arrow shown in FIG. 1, as in the periods T0 to T1. At that time, since the first rectifier diode 4b is off, the secondary current of the transformer 3 flows only through the second secondary winding 3c and increases linearly. When the first rectifier diode 4b is off, the first snubber capacitor 8Cb stores a charge proportional to the reverse bias Vb = + 2Vin / n with respect to the first rectifier diode 4b.
[0107]
As shown in FIG. 2, the primary voltage Vt maintains a substantially constant value −Vin in the periods T4 to T5. Therefore, the excitation current included in the primary current It increases substantially linearly. On the other hand, since the secondary current of the transformer 3 increases linearly as already described, the primary side converted current increases linearly. Accordingly, the primary current It increases substantially linearly.
[0108]
<Period T5-T7>
At time T5, the switching control circuit 7 changes the third switching signal G3 from H to L, and turns off the second high-side switch unit 2H. On the other hand, the first low-side switch unit 1L maintains the on state. At that time, resonance occurs in the leakage inductance of the primary winding 3a, the parasitic capacitor 2HC of the second high-side switch unit 2H, and the parasitic capacitor 2LC of the second low-side switch unit 2L. Due to the resonance, the primary current It charges the parasitic capacitor 2HC of the second high-side switch unit 2H and simultaneously discharges the parasitic capacitor 2LC of the second low-side switch unit 2L. Therefore, the voltage V2H between both ends of the second high side switch section 2H rises smoothly from 0, and the voltage V2L between both ends of the second low side switch section 2L falls smoothly from the maximum value Vin.
[0109]
At time T6, the voltage V2H across the second high-side switch unit 2H reaches the maximum value Vin. At the same time, the voltage V2L across the second low-side switch unit 2L reaches zero. At that time, the diode 2LD of the second low-side switch unit 2L is turned on, and the voltage V2L between both ends is clamped to zero. As a result, the primary voltage Vt rises smoothly and is clamped at zero.
[0110]
In each of the first secondary winding 3b and the second secondary winding 3c, the voltage between both ends rises with the rise of the primary voltage Vt in the period T5 to T6. At this time, since the second rectifier diode 4c is on, the voltage Vc between both ends is substantially zero. On the other hand, the voltage Vb across the first rectifier diode 4b drops. That is, the reverse bias with respect to the first rectifier diode 4b decreases. At that time, the first snubber capacitor 8Cb is discharged. Based on the discharge current, the first snubber capacitor 8Cb and the leakage inductance on the secondary side of the transformer 3 start to resonate. Here, the leakage inductance on the secondary side is substantially equal to the leakage inductance on the secondary side of the transformer when the primary winding is short-circuited, as in the periods T1 to T3. Furthermore, the capacity of the first snubber capacitor 8Cb is substantially equal to that of the second snubber capacitor 8Cc, and is sufficiently larger than the capacity of the parasitic capacitor in the switch section. Accordingly, the resonance period between the first snubber capacitor 8Cb and the secondary side leakage inductance is sufficiently longer than the resonance period between the parasitic capacitor in the switch section and the primary winding 3a. As a result, the voltage Vb across the first rectifier diode 4b drops more slowly than the rise of the primary voltage Vt after time T5, and reaches 0 just before time T7 after time T6. At that time, the first rectifier diode 4b is turned on.
[0111]
The first snubber capacitor 8Cb is discharged by resonance in the periods T5 to T7. The discharge current rapidly and smoothly increases the first secondary current Ib flowing through the first secondary winding 3b. As a result, the second secondary current Ic flowing through the second secondary winding 3c decreases rapidly and smoothly. As a result, there is almost no difference between the secondary currents Ib and Ic flowing through the secondary windings 3b and 3c, respectively, in the vicinity of time T7. In this manner, in the period T5 to T7, the secondary currents of the secondary windings 3b and 3c are quickly canceled out, so that the primary side converted current is rapidly and smoothly reduced to substantially zero. Therefore, the primary current It decreases rapidly and smoothly from the value at time T5.
[0112]
<Period T7 to T8>
At time T7, the switching control circuit 7 changes the second switching signal G2 from H to L, and turns off the first low-side switch unit 1L. As a result, all four switch sections are turned off. At that time, the voltages V1L and V2L across the two low-side switch sections 1L and 2L are both zero. Accordingly, the voltages V1H and V2H across the two high-side switch units 1H and 2H are equal to the input voltage Vin. After time T7, the voltage across the high-side switch section drops from Vin with ringing, and the voltage across the low-side switch section rises from 0 with ringing.
[0113]
The primary voltage Vt of the transformer 3 is substantially maintained at 0 in the periods T7 to T8. Accordingly, the induced voltages of the two secondary windings 3b and 3c of the transformer 3 are also maintained at 0. Further, both the first rectifier diode 4b and the second rectifier diode 4c are on. Therefore, the voltage V5 across the smoothing inductor 5 is maintained at a negative constant voltage −Vout. As a result, the output current I5 flowing through the smoothing inductor 5 decreases smoothly and linearly with a gentle slope.
[0114]
The primary current It of the transformer 3 once decreases rapidly in the period T5 to T7, and then decreases to 0 more rapidly at time T7. Therefore, since the decrease is slow compared with the conventional hard switching, noise such as undershoot is reduced.
[0115]
After time T7, the primary current It does not substantially flow. Therefore, the secondary current of the transformer 3 continues to flow through the two secondary windings 3b and 3c in opposite directions and substantially in equal amounts so that the primary side converted current is zero. That is, in the period T7 to T8, the first secondary current Ib and the second secondary current Ic are substantially equal.
Thus, the state immediately before time T0 is reproduced in the periods T7 to T8. Thereafter, the operation in the period T0 to T8 is repeated.
[0116]
In switching by the switching control circuit 7, a voltage conversion rate, that is, a ratio between the input voltage Vin and the output voltage Vout is obtained as follows:
The sum of the time of the first period T0 to T1 when the first high side switch unit 1H is on and the time of the second period T4 to T5 when the second high side switch unit 2H is on Is Ton. Since the voltage (Vin / n−Vout) is applied to the smoothing inductor 5 in the first period and the second period, the magnetic flux stored in the smoothing inductor 5 is increased by (Vin / n−Vout) × Ton. .
[0117]
Let Toff be the sum of the respective times of the third period in which all the switch sections are off, that is, the periods T3 to T4 and the periods T7 to T8. Since the voltage (−Vout) is applied to the smoothing inductor 5 in each period, the magnetic flux stored in the smoothing inductor 5 is reduced by Vout × Toff. Here, the time (delay time) of the fourth period in which the above resonance occurs, that is, the period T1 to T3 and the period T5 to T7, is sufficient compared to the respective times from the first period to the third period Short. Therefore, the fourth period is ignored in the following description of the voltage conversion rate.
The condition for balancing the increase and decrease of the magnetic flux of the smoothing inductor 5 in the period T0 to T8, that is, the reset condition of the smoothing inductor 5 is expressed by the following equation (1).
[0118]
(Vin / n−Vout) × Ton = Vout × Toff (1)
[0119]
From equation (1), the voltage conversion rate (ratio between input voltage Vin and output voltage Vout) is obtained by the following equation (2).
[0120]
Vout / Vin = δ / n where δ = Ton / (Ton + Toff) (2)
[0121]
As shown in Expression (2), the output voltage Vout is stably maintained at a substantially constant value by controlling the on / off time ratio δ in the high-side switch unit.
[0122]
At the end of each of the first period and the second period, the switching control circuit 7 realizes the third period across the fourth period as described above. The primary current It of the transformer 3 once decreases rapidly and smoothly in the fourth period, and further decreases rapidly to 0 at the start of the third period. Thus, the change in the primary current It when all four switch parts are OFF is more gradual than in the conventional hard switching. Therefore, the switching loss accompanying the interruption of the primary current It is reduced as compared with the conventional hard switching.
[0123]
  The time of the fourth period, ie the delay time, is preferably substantially ¼ of the resonance period determined by the equivalent capacity of the snubber and the leakage inductance on the secondary side of the transformer 3 when the primary winding 3a is short-circuited. be equivalent to. For example, as indicated by the one-dot chain line A in FIG. 2, the time during which the fourth switching signal G4 is H, that is, the delay time is extended. Thereby, the primary current It of the transformer 3 is the second current as shown by the one-dot chain line B in FIG.Snubber capacitor 8CcAnd the resonance end time T3 with the secondary side leakage inductance of the transformer 3 continues to decrease. However, the decrease in the primary current It in the third period T3 to T4 is considerably slower than that in the fourth period T1 to T3. Accordingly, after time T3, the primary current It is substantially constant. Therefore, the switching loss accompanying the interruption of the primary current It is almost constant with respect to the extension of the delay time. On the other hand, since the extension of the delay time extends the conduction time of the primary current It, the circulating current loss increases. After all, the delay time is the secondSnubber capacitor 8CcAnd the time from the start of resonance between the leakage inductance of the transformer 3 and the secondary side of the transformer 3 to the turn-on of the second rectifier diode 4c, that is, substantially equal to 1/4 of the resonance period, or adjusted to several times at most. The Thereby, the circulating current loss is effectively reduced without substantially changing the switching loss accompanying the interruption of the primary current It. Thus, the efficiency of the switching power supply device 10 according to the first embodiment is high.
[0124]
In the first embodiment, the anodes of the two rectifier diodes 4b and 4c are connected to the secondary winding and the cathode is connected to the smoothing inductor 5, respectively. Conversely, the cathodes of the two rectifier diodes 4b and 4c may be connected to the secondary winding and the anode may be connected to the smoothing inductor 5, respectively. At that time, the output voltage Vout is a negative voltage contrary to the above.
Further, separately from the above, the smoothing inductor 5 may be connected to the common end R of the two secondary windings, and one end of the two rectifier diodes 4b and 4c may be connected to the second output terminal 2b.
[0125]
Example 2
The switching power supply device according to the second embodiment of the present invention has the same configuration as that according to the first embodiment 10, and particularly includes a full bridge converter. However, unlike the first embodiment, the switching control circuit 7 performs soft switching on the four switch units 1H, 1L, 2H, and 2L.
Since the circuit configuration of the switching power supply according to the second embodiment is the same as that of the first embodiment shown in FIG. 1, FIG. 1 is used for the circuit configuration.
[0126]
In the switching control in the first embodiment, the high-side switch unit and the low-side switch unit are simultaneously turned on at the start of each of the first period and the second period, as at times T0 and T4 in FIG. At that time, the switching loss increases as follows: Each switch part in the full bridge includes a parasitic capacitor connected in parallel with the switch element as shown in FIG. When all the switch parts are off, the respective parasitic capacitors are stabilized in a state where charges are stored. Therefore, when the switch unit is turned on, the parallel switch elements are turned on while the parasitic capacitor stores a certain amount of charge. At that time, the parasitic capacitor is short-circuited through the turned-on switch element and is rapidly discharged. As a result, a surge current is generated in the switch portion and converted into heat or electromagnetic waves. Electric power is dissipated outside through the heat and electromagnetic waves. Thus, switching loss increases.
[0127]
For example, in FIG. 2, the voltages V1H and V2L between both ends of the first high-side switch unit 1H and the second low-side switch unit 2L are kept at finite values just before time T0. Accordingly, the parasitic capacitors 1HC and 2LC of the first high-side switch unit 1H and the second low-side switch unit 2L store charges proportional to the voltages V1H and V2L, respectively. When both switch parts are turned on at time T0, both parasitic capacitors 1HC and 2LC are suddenly discharged, and a surge current is generated. As a result, the current I1H flowing through the first high-side switch unit 1H and the current I2L flowing through the second low-side switch unit 2L increase so as to form a steep peak sc. When the first low-side switch unit 1L and the second high-side switch unit 2H are simultaneously turned on at time T4, similar peaks appear in the currents I1L and I2H flowing therethrough.
[0128]
In the switching according to the first embodiment, the voltage between both ends of the high-side switch unit and the low-side switch unit changes with ringing at the start of the third period. The surge voltage that causes these ringing increases the switching loss.
For example, in FIG. 2, when the second low-side switch unit 2L is turned off at time T3, the voltage V1L across the first low-side switch unit 1L and the voltage V2L across the second low-side switch unit 2L ring respectively. Ascending from 0. As a result, as shown in FIG. 2, the peaks change so as to form steep peaks sv1 and sv2, respectively. Similar peaks appear in the voltages V1L and V2H across the two low-side switch units 1L and 2L when the first low-side switch unit 1L is turned off at time T7. Therefore, switching loss increases.
[0129]
The switching power supply according to the second embodiment realizes soft switching, particularly ZVS, as follows when the low-side switch unit is turned on. Thereby, the generation of the surge current / surge voltage is suppressed, and the switching loss is further reduced. FIG. 3 is a waveform diagram of current and voltage generated in each part of the circuit shown in FIG. 1 due to switching of the switching control circuit 7 in the second embodiment. Here, the current and voltage shown in FIG. 3 are the same as those in FIG. 2, and the direction of the arrow shown in FIG. 1 is positive.
[0130]
The switching in the second embodiment realizes the following five periods alternately at a predetermined time and in a predetermined cycle:
(1) The first period corresponds to periods T0 to T1 in FIG. In the first period, the first high-side switch unit 1H and the second low-side switch unit 2L are turned on, and the second high-side switch unit 2H and the first low-side switch unit 1L are turned off.
(2) The second period corresponds to periods T4 to T5 in FIG. In the second period, the first high-side switch unit 1H and the second low-side switch unit 2L are turned off, and the second high-side switch unit 2H and the first low-side switch unit 1L are turned on.
(3) The third period corresponds to the periods T3 to T4 and the periods T7 to T8 in FIG. 3, and is realized between the first period and the second period. In the third period, only one of the first low-side switch unit 1L and the second low-side switch unit 2L is turned on, and the remaining three switch units are turned off.
[0131]
(4) The dead time is a minute period realized at the end of each of the first period and the second period, and corresponds to periods T1 to T2 and periods T5 to T6 in FIG. In the dead time, the pair of either the first high-side switch unit 1H and the first low-side switch unit 1L, or the second high-side switch unit 2H and the second low-side switch unit 2L is turned off. .
[0132]
(5) The fourth period is realized at the end of each of the first period and the second period, overlapping with the above dead time, and corresponds to periods T1 to T3 and periods T5 to T7 in FIG. In the fourth period, as in the first embodiment, resonance occurs between the snubber capacitor in the rectifying unit and the leakage inductance on the secondary side of the transformer 3. Similar to the first embodiment, the length of the fourth period is referred to as a delay time. The delay time is sufficiently longer than the dead time.
[0133]
Hereinafter, switching in the second embodiment will be described in the order of time from time T0 to time T8 shown in FIG.
<Period T0-T1>
In the periods T0 to T1, as in the periods T0 to T1 (see FIG. 2) in the first embodiment, the primary current It of the transformer 3 and the current I5 flowing through the smoothing inductor 5 both increase linearly. At that time, the first rectifier diode 4b is turned on and the second rectifier diode 4c is turned off. Therefore, the secondary current of the transformer 3 flows only through the first secondary winding 3b. Further, the second snubber capacitor 8Cc stores a charge proportional to the reverse bias Vc with respect to the second rectifier diode 4c.
[0134]
<Period T1-T2>
At time T1, the switching control circuit 7 changes the first switching signal G1 from H to L, and turns off the first high-side switch unit 1H. On the other hand, the second low-side switch unit 2L maintains the on state. At that time, resonance occurs in the leakage inductance of the primary winding 3a, the parasitic capacitor 1HC of the first high-side switch unit 1H, and the parasitic capacitor 1LC of the first low-side switch unit 1L. Due to the resonance, the primary current It charges the parasitic capacitor 1HC of the first high-side switch unit 1H and simultaneously discharges the parasitic capacitor 1LC of the first low-side switch unit 1L. Accordingly, the voltage V1H between both ends of the first high-side switch section 1H rises smoothly from 0, and the voltage V1L between both ends of the first low-side switch section 1L falls smoothly from the maximum value Vin.
[0135]
Immediately before time T2, the voltage V1H across the first high-side switch unit 1H reaches the maximum value Vin. At the same time, the voltage V1L across the first low-side switch unit 1L reaches zero. At that time, the diode 1LD of the first low-side switch unit 1L is turned on, and the voltage V1L between both ends is clamped to zero. At time T2, the switching control circuit 7 changes the second switching signal G2 from L to H, and turns on the first low-side switch unit 1L. In this way, ZVS is realized when the first low-side switch unit 1L is turned on.
[0136]
On the other hand, the voltage V2L across the second low-side switch unit 2L is maintained at 0 throughout the periods T1 to T2. Accordingly, the primary voltage Vt falls smoothly and is clamped at 0 in the same manner as the voltage V1L across the first low-side switch unit 1L.
[0137]
<Period T1-T3>
In each of the first secondary winding 3b and the second secondary winding 3c, the voltage between both ends falls with the fall of the primary voltage Vt in the period T1 to T2. At that time, since the first rectifier diode 4b is on, the voltage Vb between both ends is substantially zero. On the other hand, the voltage Vc across the second rectifier diode 4c drops. That is, the reverse bias with respect to the second rectifier diode 4c decreases. Thereby, the second snubber capacitor 8Cc is discharged. Based on the discharge current, the second snubber capacitor 8Cc and the leakage inductance on the secondary side of the transformer 3 start to resonate. Here, like the first embodiment, the leakage inductance on the secondary side is substantially equal to the leakage inductance on the secondary side of the transformer 3 when the primary winding 3a is short-circuited. Furthermore, the capacitance of the second snubber capacitor 8Cc is sufficiently larger than the capacitance of the parasitic capacitor in the switch section. Therefore, the resonance period between the second snubber capacitor 8Cc and the above-described secondary side leakage inductance is sufficiently longer than the resonance period between the parasitic capacitor in the switch section and the primary winding 3a. As a result, the voltage Vc across the second rectifier diode 4c gradually falls below the primary voltage Vt after time T1, and reaches 0 just before time T3 after time T2. At that time, the second rectifier diode 4c is turned on.
[0138]
The second snubber capacitor 8Cc is discharged by resonance in the periods T1 to T3. The discharge current rapidly and smoothly increases the second secondary current Ic flowing through the second secondary winding 3c. As a result, the first secondary current Ib flowing through the first secondary winding 3b decreases rapidly and smoothly. As a result, there is almost no difference between the secondary currents Ib and Ic flowing through the secondary windings 3b and 3c in the vicinity of time T3. In this manner, in the period T1 to T3, the secondary currents of the secondary windings 3b and 3c are quickly canceled out, so that the primary side converted current is rapidly and smoothly reduced to substantially zero. Therefore, the primary current It decreases rapidly and smoothly from the value at time T1.
[0139]
<Period T3-T4>
At time T3, the switching control circuit 7 changes the fourth switching signal G4 from H to L, and turns off the second low-side switch unit 2L. Thereby, only the first low-side switch unit 1L is turned on, and the remaining three switch units are turned off. Thus, in the second embodiment, unlike the first embodiment, the first low-side switch unit 1L is already turned on at time T2 prior to time T3. Accordingly, in the second embodiment, after time T3, the ringing of the voltage V2L across the second low-side switch unit 2L is suppressed as compared with the first embodiment, and rapidly attenuates. As a result, switching loss is reduced.
[0140]
In the period T3 to T4, the primary voltage Vt of the transformer 3 is maintained at 0. Accordingly, the induced voltages of the two secondary windings 3b and 3c of the transformer 3 are also maintained at 0. Further, both the first rectifier diode 4b and the second rectifier diode 4c are on. Therefore, the voltage V5 across the smoothing inductor 5 is stably maintained at the negative constant voltage −Vout. As a result, the output current I5 of the smoothing inductor 5 decreases smoothly and linearly with a gentle slope.
[0141]
The primary current It of the transformer 3 once decreases rapidly in the periods T1 to T3, and then decreases to 0 more rapidly at time T3. Therefore, since the decrease is slow compared with the conventional hard switching, noise such as undershoot is reduced.
After time T3, the primary current It does not substantially flow. Therefore, the secondary current of the transformer 3 continues to flow through the two secondary windings 3b and 3c in opposite directions and substantially in equal amounts so that the primary side converted current is zero. That is, in the periods T3 to T4, the first secondary current Ib and the second secondary current Ic are substantially equal.
[0142]
<Period T4-T5>
At time T4, the switching control circuit 7 changes the third switching signal G3 from L to H, and turns on the second high-side switch unit 2H. At that time, both the first high-side switch unit 1H and the second low-side switch unit 2L are turned off, and the first low-side switch unit 1L is turned on. When the first low-side switch unit 1L and the second high-side switch unit 2H are turned on, the input voltage Vin is applied to the primary winding 3a of the transformer 3 in the reverse direction during the period T0 to T1, and the primary voltage Vt is 0. Descends rapidly to -Vin. Further, the primary current It of the transformer 3 flows through the primary winding 3a from the second connection point Q to the first connection point P in the opposite direction to the periods T0 to T1. Therefore, the induced voltage of each of the two secondary windings 3b and 3c of the transformer 3 rapidly drops from 0 to −Vin / n. At that time, since the second rectifier diode 4c is already turned on, the voltage Vb across the first rectifier diode 4b rapidly rises to + 2Vin / n. Thereby, the first rectifier diode 4b is turned off. As a result, the secondary current of the transformer 3 starts to flow only through the second secondary winding 3c. Thus, commutation occurs on the secondary side of the transformer 3 at time T4.
[0143]
The voltage V5 across the smoothing inductor 5 is substantially equal to Vin / n−Vout in the periods T4 to T5, as in the periods T0 to T1. Accordingly, the output current I5 flowing through the smoothing inductor 5 increases linearly and gently in the direction of the arrow shown in FIG. 1, as in the periods T0 to T1. Since the first rectifier diode 4b is turned off, the secondary current of the transformer 3 flows only through the second secondary winding 3c and increases linearly. When the first rectifier diode 4b is off, the first snubber capacitor 8Cb accumulates a charge proportional to the reverse bias Vb = + 2Vin / n.
[0144]
As shown in FIG. 3, in the period T4 to T5, the primary voltage Vt maintains a substantially constant value −Vin. Therefore, the excitation current included in the primary current It increases substantially linearly. On the other hand, since the secondary current of the transformer 3 increases linearly as already described, the primary side converted current increases linearly. Accordingly, the primary current It increases substantially linearly.
[0145]
<Period T5-T6>
At time T5, the switching control circuit 7 changes the third switching signal G3 from H to L, and turns off the second high-side switch unit 2H. On the other hand, the first low-side switch unit 1L maintains the on state. At that time, resonance occurs in the leakage inductance of the primary winding 3a, the parasitic capacitor 2HC of the second high-side switch unit 2H, and the parasitic capacitor 2LC of the second low-side switch unit 2L. Due to the resonance, the primary current It charges the parasitic capacitor 2HC of the second high-side switch unit 2H and simultaneously discharges the parasitic capacitor 2LC of the second low-side switch unit 2L. Therefore, the voltage V2H between both ends of the second high side switch section 2H rises smoothly from 0, and the voltage V2L between both ends of the second low side switch section 2L falls smoothly from the maximum value Vin.
[0146]
Immediately before time T6, the voltage V2H between both ends of the second high-side switch unit 2H reaches the maximum value Vin. At the same time, the voltage V2L across the second low-side switch unit 2L reaches zero. At that time, the diode 2LD of the second low-side switch unit 2L is turned on, and the voltage V2L between both ends is clamped to zero. At time T6, the switching control circuit 7 changes the fourth switching signal G4 from L to H, and turns on the second low-side switch unit 2L. In this way, ZVS is realized when the second low-side switch unit 2L is turned on.
The voltage V1L across the first low-side switch unit 1L is maintained at 0 throughout the periods T5 to T7. Accordingly, the primary voltage Vt rises smoothly like the voltage V2L across the second low-side switch unit 2L, and is clamped at 0.
[0147]
<Period T5-T7>
In each of the first secondary winding 3b and the second secondary winding 3c, the voltage between both ends rises with the rise of the primary voltage Vt in the period T5 to T6. At this time, since the second rectifier diode 4c is on, the voltage Vc between both ends is substantially zero. On the other hand, the voltage Vb across the first rectifier diode 4b drops. That is, the reverse bias with respect to the first rectifier diode 4b decreases. Thereby, the first snubber capacitor 8Cb is discharged. Based on the discharge current, the first snubber capacitor 8Cb and the leakage inductance on the secondary side of the transformer 3 start to resonate. Here, the leakage inductance on the secondary side is substantially equal to the leakage inductance on the secondary side of the transformer 3 when the primary winding 3a is short-circuited, as in the periods T1 to T3. The capacity of the first snubber capacitor 8Cb is sufficiently larger than the capacity of the parasitic capacitor in the switch section. Accordingly, the resonance period between the first snubber capacitor 8Cb and the secondary side leakage inductance is sufficiently longer than the resonance period between the parasitic capacitor in the switch section and the primary winding 3a. As a result, the voltage Vb across the first rectifier diode 4b drops more slowly than the rise of the primary voltage Vt after time T5, and reaches 0 just before time T7 after time T6. At that time, the first rectifier diode 4b is turned on.
[0148]
The first snubber capacitor 8Cb is discharged by resonance in the periods T5 to T7. The discharge current rapidly and smoothly increases the first secondary current Ib flowing through the first secondary winding 3b. As a result, the second secondary current Ic flowing through the second secondary winding 3c decreases rapidly and smoothly. As a result, there is almost no difference between the secondary currents Ib and Ic flowing through the secondary windings 3b and 3c, respectively, in the vicinity of time T7. In this manner, in the period T5 to T7, the secondary currents of the secondary windings 3b and 3c are quickly canceled out, so that the primary side converted current is rapidly and smoothly reduced to substantially zero. Therefore, the primary current It decreases rapidly and smoothly from the value at time T5.
[0149]
<Period T7 to T8>
At time T7, the switching control circuit 7 changes the second switching signal G2 from H to L, and turns off the first low-side switch unit 1L. Thereby, only the second low-side switch unit 2L is turned on, and the remaining three switch units are turned off. Thus, unlike the first embodiment, the second low-side switch unit 2L is already turned on at the time T6 prior to the time T7 in the second embodiment. Therefore, in the second embodiment, after time T7, the ringing of the voltage V2L across the second low-side switch unit 2L is suppressed as compared with the first embodiment and rapidly attenuates. As a result, switching loss is reduced.
[0150]
In the periods T7 to T8, the primary voltage Vt of the transformer 3 is maintained at 0. Accordingly, the induced voltages of the two secondary windings 3b and 3c of the transformer 3 are also maintained at 0. Further, both the first rectifier diode 4b and the second rectifier diode 4c are on. Therefore, the voltage V5 across the smoothing inductor 5 is stably maintained at the negative constant voltage −Vout. As a result, the output current I5 of the smoothing inductor 5 decreases smoothly and linearly with a gentle slope.
[0151]
The primary current It of the transformer 3 once decreases rapidly in the period T5 to T7, and then decreases to 0 more rapidly at time T7. Therefore, since the decrease is slow compared with the conventional hard switching, noise such as undershoot is reduced.
[0152]
After time T7, the primary current It does not substantially flow. Therefore, the secondary current of the transformer 3 continues to flow through the two secondary windings 3b and 3c in opposite directions and substantially in equal amounts so that the primary side converted current is zero. That is, in the period T7 to T8, the first secondary current Ib and the second secondary current Ic are substantially equal.
Thus, the state immediately before time T0 is reproduced in the periods T7 to T8. Thereafter, the operation in the period T0 to T8 is repeated.
[0153]
In the switching according to the second embodiment, the voltage conversion rate (ratio between the input voltage Vin and the output voltage Vout) is obtained as in the first embodiment as follows:
The fourth period in which resonance occurs on the secondary side of the transformer 3, that is, the time (delay time) of each of the periods T1 to T3 and the periods T5 to T7, is the respective period from the first period to the third period. Short enough compared to time. Therefore, as in the first embodiment, the fourth period can be ignored from the first period to the third period. In the first period T0 to T1 and the second period T4 to T5, the magnetic flux stored in the smoothing inductor 5 is increased by (Vin / n−Vout) × Ton. Here, the sum of the time of the first period T0 to T1 and the time of the second period T4 to T5 is Ton. On the other hand, in the third period, that is, the period T3 to T4 and the period T7 to T8, the magnetic flux stored in the smoothing inductor 5 is reduced by Vout × Toff. Here, the time of the third period is Toff. Accordingly, the reset condition of the smoothing inductor 5 in the period T0 to T8 is expressed by the following equation (1), as in the first embodiment:
[0154]
(Vin / n−Vout) × Ton = Vout × Toff (1)
[0155]
From the equation (1), the voltage conversion rate is obtained by the following equation (2) as in the first embodiment.
[0156]
Vout / Vin = δ / n where δ = Ton / (Ton + Toff) (2)
[0157]
As shown in Expression (2), the output voltage Vout is stably maintained at a substantially constant value by controlling the on / off time ratio δ in the high-side switch unit.
[0158]
At the end of each of the first period and the second period, the switching control circuit 7 realizes the third period across the fourth period as described above. The primary current It of the transformer 3 once decreases rapidly and smoothly in the fourth period, and further decreases rapidly to 0 at the start of the third period. Thus, the change when the primary current It is cut off is more gradual than the conventional hard switching. Therefore, the switching loss accompanying the interruption of the primary current It is reduced as compared with the conventional hard switching.
[0159]
Further, in the second embodiment, unlike the first embodiment, at the start of the third period, the ringing of the voltage across one low-side switch unit is suppressed by turning on the other low-side switch unit and rapidly attenuates. . Therefore, no substantial surge voltage is generated. Thus, the switching power supply device according to the second embodiment further reduces the switching loss at the start of the third period.
[0160]
  The time of the fourth period, that is, the delay time is preferably the same as that of the first embodiment, preferably the resonance period determined by the equivalent capacity of the snubber and the leakage inductance on the secondary side of the transformer 3 when the primary winding 3a is short-circuited. Substantially equal to 1/4. For example, as indicated by the one-dot chain line A in FIG. 3, the time during which the fourth switching signal G4 is H, that is, the delay time is extended. As a result, the primary current It of the transformer 3 is the second current as shown by the one-dot chain line B in FIG.Snubber capacitor 8CcAnd the resonance end time T3 with the secondary side leakage inductance of the transformer 3 continues to decrease. However, after time T3, the decrease in the primary current It in the third period T3 to T4 is considerably slower than in the fourth period T1 to T3. Therefore, the primary current It is substantially constant after time T3. Therefore, the switching loss accompanying the interruption of the primary current It is almost constant with respect to the extension of the delay time. On the other hand, since the extension of the delay time extends the conduction time of the primary current It, the circulating current loss increases. After all, the delay time is the secondSnubber capacitor 8CcAnd the time from the start of resonance between the leakage inductance of the transformer 3 and the secondary side of the transformer 3 to the turn-on of the second rectifier diode 4c, that is, substantially equal to 1/4 of the resonance period, or adjusted to several times at most. The Thereby, the circulating current loss is effectively reduced without substantially changing the switching loss accompanying the interruption of the primary current It. Thus, the efficiency of the switching power supply according to the second embodiment is high.
[0161]
In the second embodiment, the two low-side switch units 1L and 2L are turned on in any of the periods T2 to T3 and the periods T6 to T7. Thereby, the primary current It of the transformer 3 circulates only in the closed circuit composed of the primary winding 3a and the two low-side switch parts 1L and 2L. In addition, the two high-side switch parts 1H and 2H are turned on in both or any of the above periods, and the primary current It is circulated through a closed circuit composed of the primary winding 3a and the two high-side switch parts 1H and 2H. You may let them. In particular, the closed circuit on the low side may be alternately circulated in the periods T2 to T3, and the closed circuit on the high side may be alternately circulated in the periods T6 to T7. At that time, the full bridge circuit including the four switch units receives the stress evenly from the circulation of the primary current It. This avoids concentration of stress on any of the switch sections.
[0162]
In the second embodiment, as in the first embodiment, the polarities of the two rectifier diodes 4b and 4c may be reversed. At that time, the output voltage Vout is a negative voltage contrary to the above.
Further, separately from the above, the smoothing inductor 5 may be connected to the common end R of the two secondary windings, and one end of the two rectifier diodes 4b and 4c may be connected to the second output terminal 2b.
[0163]
Example 3
FIG. 4 is a diagram illustrating a circuit of the switching power supply device 10a according to the third embodiment of the present invention. The switching power supply device 10a according to the third embodiment is different from the switching power supply device 10 according to the first embodiment shown in FIG. 1 in the snubbers 8b and 8c. Since others have the same configuration as in the first embodiment, the same components as those in FIG. 1 are denoted by the same reference numerals, and the description of the first embodiment is incorporated.
[0164]
The first snubber 8b includes a first snubber resistor 8Rb in series with the first snubber capacitor 8Cb, and the second snubber 8c includes a second snubber resistor 8Rc in series with the second snubber capacitor 8Cc. That is, the rectifying unit includes an RC snubber in parallel with each of the rectifying diodes 4b and 4c.
[0165]
In Example 1 and Example 2 described above, surge current / surge voltage is generated when the rectifier diodes 4b and 4c are turned off at the time of commutation on the secondary side. As a result, as shown in FIGS. 2 and 3, respectively, a peak cf is generated in the secondary currents Ib and Ic flowing through the secondary windings 4b and 4c of the transformer 3, respectively, and both ends of the two rectifier diodes 4b and 4c. A peak vf occurs in the inter-voltages Vb and Vc.
In Example 3, the above-described surge current / surge voltage is reduced by the snubber resistors 8Rb and 8Rc in the snubbers 8b and 8c as follows:
[0166]
At the start of each of the first period and the second period, one of the two rectifier diodes 4b and 4c is turned off. The start of the first period and the second period corresponds to times T0 and T4 in FIGS. The diode accumulates charge while on. When the diode is turned off by sudden application of reverse bias, the accumulated charge is discharged as a reverse current. In the above-described snubbers 8b and 8c, at the start of each of the first period and the second period, the respective snubber capacitors 8Cb and 8Cc are discharged in the reverse direction with respect to the rectifier diodes 4b and 4c. Based on the discharge current, the snubber capacitors 8Cb and 8Cc resonate with the leakage inductances of the secondary windings 3b and 3c of the transformer 3, respectively. Since the capacitances of the snubber capacitors 8Cb and 8Cc are sufficiently larger than the parasitic capacitances of the rectifying diodes 4b and 4c in the same rectifying unit, the resonance period is long. Further, the peak of the resonance current is suppressed by the snubber resistors 8Rb and 8Rc. Therefore, the surge current / surge voltage associated with the resonance is suppressed.
[0167]
For example, when the reverse bias Vc is applied to the second rectifier diode 4c at time T0, the second snubber capacitor 8Cc and the leakage inductance of the second secondary winding 3c resonate. Thereby, a surge voltage is generated between both ends of the second rectifying unit. As a result, as shown in FIG. 2 and FIG. 3, the peak vf due to the surge voltage in the voltage Vc across the second rectifier diode 4c, the peak cf due to the surge current in the second secondary current Ic, Each appears. In the third embodiment, the respective peaks vf and cf are suppressed to be small by the second snubber resistance 8Rc. Similarly, at time T4, the peak due to the surge voltage / surge current is kept small by the snubber resistor 8Rb in the first snubber 8b. Thus, the third embodiment reduces the switching loss due to the rectifier diode being turned off.
[0168]
Example 4
FIG. 5 is a diagram showing a circuit of the switching power supply device 10b according to the fourth embodiment of the present invention. The switching power supply device 10b according to the fourth embodiment differs from the switching power supply device 10a according to the third embodiment shown in FIG. 4 in the snubbers 8B and 8C. Since others have the same configuration as in the third embodiment, the same components as those in FIG. 4 are denoted by the same reference numerals, and the descriptions of the first and third embodiments are incorporated.
[0169]
The first snubber 8B includes a first snubber resistor 8Rb in series with the first snubber capacitor 8Cb, and further includes a first sub-rectifier diode 8Db in parallel with the first snubber resistor 8Rb. Similarly, the second snubber 8C includes a second snubber resistor 8Rc in series with the second snubber capacitor 8Cc, and further includes a second sub-rectifier diode 8Dc in parallel with the second snubber resistor 8Rc. That is, the two rectifiers each include an RCD snubber in parallel with the rectifier diodes 4b and 4c. Here, the sub-rectifier diodes 8Db and 8Dc are connected in the same direction as the corresponding rectifier diodes 4b and 4c, respectively. For example, in the first snubber 8B, as shown in FIG. 5, since the first snubber resistor 8Rb is connected to the cathode of the first rectifier diode 4b, the cathode of the first sub-rectifier diode 8Db is the first To the cathode of the rectifier diode 4b. Conversely, when the first snubber resistor 8Rb is connected to the anode of the first rectifier diode 4b, the anode of the first sub-rectifier diode 8Db is connected to the anode of the first rectifier diode 4b. Similarly, in the second snubber 8C, the polarity of the connection terminal between the second rectifier diode 4c and the second sub-rectifier diode 8Dc is determined by the position of the second snubber resistor 8Rc with respect to the second rectifier diode 4c.
[0170]
In the fourth embodiment, similarly to the third embodiment, when the commutation on the secondary side is performed, the surge current / surge voltage caused by the turn-off of the rectifier diodes 4b and 4c are caused by the snubber resistors 8Rb and 8Rc in the snubbers 8b and 8c Reduced. In the fourth embodiment, unlike the third embodiment, the discharge current from the snubber capacitors 8Cb and 8Cc passes through the sub-rectifier diodes 8Db and 8Dc in the fourth period, and bypasses the snubber resistors 8Rb and 8Rc. Thereby, reduction of the discharge current due to the snubber resistors 8Rb and 8Rc is avoided. Accordingly, since the primary side converted current is sufficiently canceled in the fourth period, the primary current It of the transformer 3 is sufficiently reduced in the fourth period regardless of the snubber resistances 8Rb and 8Rc. Furthermore, since the resistance values of the snubber resistors 8Rb and 8Rc can be sufficiently increased, the surge current / surge voltage is effectively reduced at the start of each of the first period and the second period.
[0171]
In the above embodiment, the rectifying element in the rectifying unit is a diode. In addition, a semiconductor switch element such as an IGBT may be used. At that time, the switching is controlled by the switching control circuit in synchronism with the switching of the switch section constituting the full bridge on the primary side.
[0172]
【The invention's effect】
As described above, the switching power supply device according to the present invention includes a snubber in parallel with the rectifying element in the rectifying unit. Furthermore, when the primary current of the transformer is cut off, one of the high-side switch unit and the low-side switch unit that are turned on is turned off with a delay time from the other. Thereby, the capacity of the snubber and the leakage inductance on the secondary side of the transformer resonate. As a result, the primary-side converted current is quickly and sufficiently canceled with the delay time, so that the primary current is temporarily reduced with the delay time. Thus, the switching current is small because the primary current is sufficiently small when cut off. Furthermore, since the primary current is cut off while the current flowing through the smoothing inductor is reduced, the circulating current loss is small. Therefore, the efficiency of the switching power supply device according to the present invention is high.
[Brief description of the drawings]
FIG. 1 is a circuit diagram of a switching power supply device 10 according to a first embodiment of the present invention.
2 is a waveform diagram of current and voltage generated in each part of the circuit shown in FIG. 1 due to switching in the switching power supply apparatus 10 according to the first embodiment. FIG. .
3 is a waveform diagram of current and voltage generated in each part of the circuit shown in FIG. 1 due to switching in the switching power supply device according to Embodiment 2. FIG.
FIG. 4 is a circuit diagram of a switching power supply device 10a according to a third embodiment of the present invention.
FIG. 5 is a circuit diagram of a switching power supply device 10b according to a fourth embodiment of the present invention.
FIG. 6 is a circuit diagram of a conventional switching power supply device 100 based on hard switching.
7 is a waveform diagram of current and voltage generated in each part of the circuit shown in FIG. 6 due to hard switching in the conventional switching power supply apparatus 100. FIG.
FIG. 8 is a circuit diagram of a conventional switching power supply device using soft switching.
9 is a waveform diagram of current and voltage generated in each part of the circuit shown in FIG. 8 due to soft switching in the conventional switching power supply apparatus 110. FIG.
[Explanation of symbols]
10 Switching power supply
1H First high-side switch
1L first low-side switch
2H Second high-side switch
2L Second low side switch
3 transformer
3a Primary winding
3b First secondary winding
3c Second secondary winding
4b First rectifier diode
4c Second rectifier diode
5 Smoothing inductor
6 Smoothing capacitor
8Cb first snubber capacitor
8Cc second snubber capacitor

Claims (1)

A four switching portions, any of which, a switching element is turned on and off by the switching signal from the outside, and a diode connected in parallel to the switching element, a capacitor connected in parallel to said switching element, Including a first high side switch part, a second high side switch part, a first low side switch part and a second low side switch part;
A transformer including a primary winding and a first secondary winding and a second secondary winding connected in series to each other and having the connection point as a common end;
A rectifying element, wherein including the rectifier device and the snubber including the connected snubber capacitor in parallel, respectively, the first rectifying portion and second rectifying portion; and,
A smoothing unit including a first input terminal and a second input terminal, and smoothing and outputting the input through the input terminals;
A DC-DC converter having a,
The cathode of the first high-side switch unit is substantially connected to the high potential terminal of the DC constant voltage source, the anode of the first high-side switch unit is connected to the cathode of the first low-side switch unit, and the first The anode of the low side switch unit is connected to the low potential terminal of the substantial DC constant voltage source ,
The cathode of the second high side switch unit is connected to the high potential terminal of the substantially DC constant voltage source, the anode of the second high side switch unit is connected to the cathode of the second low side switch unit, and the second The anodes of the two low-side switch sections are respectively connected to the low potential terminals of the substantial DC constant voltage source ;
To the connection point between the transformer the primary winding one end said first high-side switch unit and the first low-side switch unit, the second low-side switch and the other end the second high-side switch unit Connected to the connection point with the
To the common end another end of the first rectifier unit end and the first secondary winding of said transformer, said another end and said common end of said second secondary winding To the one end of the second rectification unit having the same polarity as the one end of the first rectification unit, the common end is connected to the first input terminal of the smoothing unit, respectively .
Each of the other end of said second rectifying portion and the first rectifier is connected to the second input terminal of the smoothing part,
DC-DC converter;
And
And equivalent capacitance of said first rectifier and said second each of the snubber of the rectification section, based on the resonance period determined by the leakage inductance of the transformer secondary at the time of short-circuit of the primary winding Determine the delay time ,
Outputting the switching signal to each of the four switch sections at a predetermined switching frequency and phase, thereby
It said first high-side switch unit and the second low-side switch unit is turned on, and the second high-side switch unit and the first low-side switch unit and the first period of time to turn off, as well as,
Wherein the first high-side switch unit and the second off the low-side switch unit and the second high-side switch unit and the first low-side switch unit and the second period to turn on,
Are alternately realized at a predetermined time and a predetermined cycle ,
At the end of the first period, one of said first of said second low-side switch unit and the high-side switch portion by turning off before the other, short-circuiting the primary winding of the transformer The equivalent capacity of the snubber is resonated with the leakage inductance on the secondary side of the transformer in the first rectification unit or the second rectification unit in which the rectification element is turned off, and the delay time elapses. By turning off the first high-side switch unit or the second low-side switch unit that is turned on, the current flowing through the primary winding of the transformer is interrupted ,
At the end of the second period, one of said second of said first low-side switch unit and the high-side switch portion by turning off before the other, short-circuiting the primary winding of the transformer The equivalent capacity of the snubber is resonated with the leakage inductance on the secondary side of the transformer in the first rectification unit or the second rectification unit in which the rectification element is turned off, and the delay time elapses. When turning off the second high-side switch part or the first low-side switch part that is turned on, the current flowing through the primary winding of the transformer is interrupted,
Switching controller for;
A switching power supply device comprising:

Images