DC conversion apparatus
Granted 27 Mar 2007 · no office action yet
Assignee: Sanken Electric Co., Ltd.
Law firm: Law firm · Log in to unlock
Attorney: Attorney · Log in to unlock
Inventors: Shinji Aso, Mamoru Tsuruya, Makoto Sato · Examiner: Gary L Laxton · AU 2838 · TC 2800
Life of the application
6 dated eventsAbstract
A DC conversion apparatus includes a switch converting a DC voltage of a DC power source to a high-frequency voltage by being turned ON/OFF via a primary winding of a transformer, a series circuit of a switch and a clamp capacitor being connected to both ends of the primary winding, a synchronous rectifying circuit performing synchronous rectification of a high-frequency voltage generated at a secondary winding, a smoothing circuit smoothing a rectified output of the synchronous rectifying circuit by a smoothing reactor and a smoothing capacitor, a capacitor storing energy stored in the smoothing reactor due to a current backflow of the synchronous rectifying circuit caused under light load condition, a switch being turned ON in synchronization with the switch and returning the energy stored in the capacitor to the clamp capacitor on the primary side via the transformer, and a control circuit turning the switch and the switch ON/OFF states alternately.
Description
10 parts›BACKGROUND OF THE INVENTION
The present invention relates to a highly efficient and compact DC conversion apparatus such as a DC/DC converter.
FIG. 1 is a circuit configuration diagram of a conventional DC conversion apparatus. In the DC conversion apparatus shown in FIG. 1 , a synchronous rectifier that includes a low ON-resistance power MOSFET (Field Effect Transistor) and the like is used on a secondary side (output side) of a transformer in order to reduce power losses.
In FIG. 1 , a MOSFET switch Q 1 is connected to a DC power source Vdc 1 via a primary winding 5 a (the number of turns: n 1 ) of a transformer T 1 , and a series circuit of a resistor R 1 and a capacitor C 1 is connected to both ends of the switch Q 1 . A series circuit of a diode D 2 and a capacitor C 2 is connected to both ends of the primary winding 5 a of the transformer T 1 , and a resistor R 2 is connected to both ends of the capacitor C 2 . The switch Q 1 is turned ON/OFF by PWM control of a control circuit 10 .
The primary winding 5 a of the transformer T 1 and a secondary winding 5 b (with a number of turns of n 2 ) of the transformer T 1 each have a winding structure such that a common mode voltage is generated mutually, and a MOSFET switch Q 3 and a MOSFET switch Q 4 are connected in series to both ends of the secondary winding 5 b of the transformer T 1 . One end of the secondary winding 5 b (filled circle (∘) side) of the transformer T 1 is connected to a gate of the switch Q 4 , and the other end thereof is connected to a gate of the switch Q 3 . The switch Q 3 is connected to a diode D 3 in parallel, and the switch Q 4 is connected to a diode D 4 in parallel. These devices form a synchronous rectifying circuit. This synchronous rectifying circuit rectifies a voltage (i.e. ON/OFF-controlled pulse voltage) generated at the secondary winding 5 b of the transformer T 1 in synchronization with ON/OFF operations of the switch Q 1 , and outputs a DC voltage.
A smoothing reactor Lo and a smoothing capacitor Co are connected in series to both ends of the switch Q 3 to form a smoothing circuit. This smoothing circuit smoothes a rectified output of the synchronous rectifying circuit, and outputs a DC output to a load RL.
The control circuit 10 controls the switch Q 1 in ON/OFF manner so that the width of an ON-pulse applied to the switch Q 1 is narrowed when the output voltage of the load RL reaches or exceeds a reference voltage. That is, when the output voltage of the load RL reaches or exceeds the reference voltage, the width of the ON-pulse applied to the switch Q 1 is shortened so as to control the output voltage to a constant voltage.
Operations of the DC conversion apparatus thus configured are explained with reference to a timing chart at light load time shown in FIG. 2 . In FIG. 2 , a drain-source voltage of the switch Q 1 is depicted by Q 1 v , a drain current of the switch Q 1 is depicted by Q 1 i , a drain current of the switch Q 3 is depicted by Q 3 i , a drain current of the switch Q 4 is depicted by Q 4 i , a drain-source voltage of the switch Q 3 is depicted by Q 3 v , and a gate voltage signal of the switch Q 1 is depicted by Q 1 g.
Operations performed at heavy load time are explained first. When the switch Q 1 is turned ON by the gate voltage signal Q g , the current Q 1 i flows through a path passing along Vdc 1 , 5 a, Q 1 , and Vdc 1 . This current Q 1 i increases linearly as time passes.
At this time, since a voltage is generated also at the secondary winding 5 b of the transformer T 1 , the switch Q 4 is turned ON and hence the current Q 4 i flows clockwise through a path passing along 5 b, Lo, Co, Q 4 , and 5 b, so that electric power is supplied to the load RL. The current Q 4 i increases linearly as time passes, whereupon Lo (Io) 2/2 of energy is stored in the smoothing reactor Lo. A current flowing into the smoothing rector Lo is depicted by Io.
Next, when the switch Q 1 is turned OFF, the voltage at the secondary winding 5 b of the transformer T 1 is reversed, and hence the switch Q 4 is turned OFF and the switch Q 3 is turned ON. Therefore, the energy stored in the smoothing reactor Lo causes the current Q 3 i to flow clockwise through a path passing along Lo, Co, Q 3 , and Lo, so that electric power is supplied continuously to the load RL.
Subsequently, when the switch Q 1 is turned ON, the voltage generated at the secondary winding 5 b is reversed again, therefore the switch Q 4 is turned ON and the switch Q 3 is turned OFF, and then the similar operations as described above are performed. This state is called a continuous mode because the current of the smoothing reactor Lo flows continuously in the same direction.
On the other hand, when a load current decreases (in a case of light load), the current of the smoothing reactor Lo flowing therethrough as the switch Q 1 is turned OFF (e.g., time t 32 ) becomes zero while the switch Q 1 is OFF, but the switch Q 3 remains ON. Therefore, the electric charge stored in the smoothing capacitor Co is discharged, and then a current Q 3 i ′ flows counterclockwise through a path passing along Co, Lo, Q 3 , and Co, so that the energy is stored in the smoothing reactor Lo.
When the switch Q 1 is turned ON at time t 33 (same as time t 31 ), the switch Q 4 is turned ON and the switch Q 3 is turned OFF. Therefore, a current Q 4 i ′ flows counterclockwise through a path passing along Lo, 5 b, Q 4 , Co, and Lo from the smoothing reactor Lo. As a result, the energy is finally returned to the DC power source Vdc 1 on a primary side (input side) via the primary winding 5 a of the transformer T 1 .
As described above, when a synchronous rectifying circuit is applied to the conventional DC conversion apparatus shown in FIG. 1 or to a switching power supply device described in Patent Document 1, they operate with little loss in a heavy load state in which a current flows continuously through the smoothing reactor Lo.
A related art of the conventional switching power supply device is disclosed in, for example, Japanese Patent Application Laid-Open No. 2002-10636.
›SUMMARY OF THE INVENTION
In a light load state as shown in FIG. 2 , however, a current of the smoothing reactor Lo flowing therethrough is not continuous and flows backward in a return mode. When the switch Q 3 is OFF, the energy stored in the smoothing reactor Lo is returned to the DC power source Vdc 1 on the input side via the switch Q 4 and the transformer T 1 .
At this time, delay of an ON-timing of the switch Q 4 or high leakage inductance of the transformer T 1 generates a large spike voltage SP as shown in FIG. 2 , causing a breakdown of devices (synchronous rectifier).
In order to solve this problem, a spike voltage absorbing circuit such as an absorber having a combination of the resistor R 1 and the capacitor C 1 , and a combination of the resistor R 2 and the capacitor C 2 should be additionally provided with the apparatus. Furthermore, the reverse current is detected to stop the synchronous rectification, or high withstand voltage devices are used. These measures lead to complex circuitry and increased losses.
According to the present invention, a DC conversion apparatus that offers high efficiency and achieve a size reduction by reducing losses and simplifying the circuitry can be provided.
The present invention adopts following means to solve the above problems. A first technical aspect of the present invention provides a DC conversion apparatus comprising: a first switch configured to carry out ON/OFF operations via a primary winding of a transformer to convert a DC voltage provided by a DC power source into a high-frequency voltage; a series circuit of a second switch and a clamp capacitor, the series circuit being connected to both ends of the primary winding of the transformer or to both ends of the first switch; a synchronous rectifying circuit configured to carry out synchronous rectification of a high-frequency voltage generated at a secondary winding of the transformer; a smoothing circuit having a smoothing reactor and a smoothing capacitor configured to smooth a rectified output of the synchronous rectifying circuit and outputs a DC voltage; a capacitor configured to storage energy being stored in the smoothing reactor and transferred thereto via a diode due to a current backflow of the synchronous rectifying circuit under light load condition; a third switch configured to turn ON in synchronization with the second switch and to return the energy stored in the capacitor to the clamp capacitor on the primary side via the transformer; and a control circuit turning the first switch and the second switch ON/OFF states alternately.
A second technical aspect of the present invention provides the DC conversion apparatus, in addition to the first technical aspect, wherein the smoothing reactor is formed on an auxiliary winding of the transformer, the secondary winding of the transformer is tightly coupled with the primary winding of the transformer, and the auxiliary winding of the transformer is loosely coupled with the primary winding.
A third technical aspect of the present invention provides the DC conversion apparatus, in addition to the second technical aspect, wherein the number of turns of the secondary winding of the transformer is the same as that of the auxiliary winding, the secondary winding has a winding phase opposite to that of the primary winding of the transformer, and the auxiliary winding has the same winding phase as that of the primary winding.
A fourth technical aspect of the present invention provides the DC conversion apparatus, in addition to the first to the fourth technical aspects, further comprising a tertiary winding of the transformer being connected to the secondary winding of the transformer in series, wherein the third switch is turned ON by a voltage generated at the tertiary winding of the transformer as the second switch is turned ON and the energy stored in the capacitor is returned to the clamp capacitor on the primary side via the transformer.
A fifth technical aspect of the present invention provides the DC conversion apparatus, in addition to the fourth technical aspect, wherein the tertiary winding of the transformer has a winding phase opposite to that of the primary winding of the transformer.
A sixth technical aspect of the present invention provides the DC conversion apparatus, in addition to the first to the fifth technical aspects, further comprising an ON-time setting means setting an ON-time of the third switch to be shorter than that of the second switch.
A seventh technical aspect of the present invention provides the DC conversion apparatus, in addition to the first to the sixth technical aspects, wherein the synchronous rectifying circuit has a parallel circuit of a synchronous rectifying device and a rectification device being connected to one end of the secondary winding of the transformer and to one end of the smoothing capacitor, and also has a second diode being connected to a joint of the one end of the secondary winding and a cathode of the rectification device, and to one end of the capacitor, and upon recovery phase of the rectification device, the recovery energy is absorbed in the capacitor via the second diode.
›BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a circuit configuration diagram of a conventional DC conversion apparatus;
FIG. 2 is a timing chart of signal of each component under light load of the DC conversion apparatus shown in FIG. 1 ;
FIG. 3 is a circuit configuration diagram of a DC conversion apparatus according to a first embodiment;
FIG. 4 is a diagram of an example of a waveform shaping circuit provided in the DC conversion apparatus according to the first embodiment;
FIG. 5 is a diagram of an operational waveform of the waveform shaping circuit shown in FIG. 4 ;
FIG. 6 is a signal timing chart of each component under light load of the DC conversion apparatus according to the first embodiment;
FIG. 7 is a timing chart of signals at each component under heavy load of the DC conversion apparatus when only a synchronous rectifying circuit is provided;
FIG. 8 is a timing chart of signal of each component under heavy load of the DC conversion apparatus according to the first embodiment;
FIG. 9 is a circuit configuration diagram of a DC conversion apparatus according to a second embodiment;
FIG. 10 is a circuit configuration diagram of a DC conversion apparatus according to a third embodiment;
FIG. 11 is a configuration diagram of a transformer provided in the DC conversion apparatus according to the third embodiment;
FIG. 12 is a signal timing chart of each component of the DC conversion apparatus according to the third embodiment; and
FIG. 13 is a circuit configuration diagram of a DC conversion apparatus according to a fourth embodiment.
›DESCRIPTION OF THE PREFERRED EMBODIMENTS · 1 of 5
Embodiments of a DC conversion apparatus according to the present invention will be explained in detail below with reference to the accompanying drawings. According to the DC conversion apparatus of the embodiments, a spike voltage is eliminated in such that energy stored in a smoothing reactor is transferred to a capacitor via a diode utilizing a current backflow caused under light load condition, and the energy is returned to a clamp capacitor on a primary side without any power loss by use of a third switch which is turned ON synchronously as a second switch is turned ON. Furthermore, the DC conversion apparatus of the embodiments suppresses a spike voltage caused by recovery of rectifying devices provided in a synchronous rectifying circuit.
First Embodiment
FIG. 3 is a circuit configuration diagram of a DC conversion apparatus in a first embodiment shown in FIG. 3 , a MOSFET switch Q 1 (first switch) is connected to a DC power source Vdc 1 via a primary winding 5 a (the number of turns is n 1 ) of a transformer T 2 , and a diode D 1 and a capacitor C 1 are connected in parallel to both ends of the switch Q 1 . The diode D 1 can be a parasitic diode of the switch Q 1 , and the capacitor C 1 can have a parasitic capacity of the switch Q 1 . One end of a MOSFET switch Q 2 (second switch) is connected to a joint of one end of the primary winding 5 a of the transformer T 2 and one end of the switch Q 1 , and the other end of the switch Q 2 is connected to a positive electrode of the DC power source Vdc 1 via a clamp capacitor C 2 . Alternatively, the other end of the switch Q 2 can be connected to a negative electrode of the DC power source Vdc 1 via the clamp capacitor C 2 .
A MOSFET switch Q 3 and a MOSFET switch Q 4 are connected in series to both ends of a secondary winding 5 b of the transformer T 2 . One end (filled circle side) of the secondary winding 5 b of the transformer T 2 is connected to a gate of the switch Q 4 , and the other end thereof is connected to a gate of the switch Q 3 . The switch Q 3 is connected to a diode D 3 in parallel, and the switch Q 4 is connected to a diode D 4 in parallel. The configuration of these devices performs a synchronous rectifier. The synchronous rectifying circuit rectifies a voltage (ON/OFF-controlled pulse voltage) generated at the secondary winding 5 b of the transformer T 2 by ON/OFF operations of the switches Q 3 and Q 4 performed in synchronization with ON/OFF operations of the switch Q 1 , and outputs a DC voltage.
A diode D 2 is connected in parallel to both ends of the switch Q 2 . The diode D 2 can be a parasitic diode of the switch Q 2 . The switches Q 1 and Q 2 have common periods being turned OFF (dead time), and are alternately turned ON/OFF by PWM control of a control circuit 10 a.
The transformer T 2 has the primary winding 5 a, the secondary winding 5 b (the number of turns is n 2 ) that is tightly coupled with the primary winding 5 a and is configured to generate a common mode voltage, and a tertiary winding 5 c (the number of turns is n 3 ) that is tightly coupled with the primary winding 5 a and is configured to generate a reverse mode voltage. The secondary winding 5 b and the tertiary winding 5 c are connected in series.
The reverse mode voltage of the tertiary winding 5 c described herein means that the tertiary winding 5 c has a winding structure with respect to the primary winding 5 a configured to turn OFF as the switch Q 1 being ON and turn ON as the switch Q 1 being OFF.
The anode of a diode D 5 is connected to one end of the secondary winding 5 b, the cathode of the diode D 3 , and one end of the smoothing reactor Lo. The cathode of the diode D 5 is connected to one end of a capacitor C 5 , and the other end of the capacitor C 5 is connected to one end of the smoothing capacitor Co and the anodes of the diodes D 3 and D 4 . A joint of the cathode of the diode D 5 and one end of the capacitor C 5 is connected to one end (drain) of a MOSFET switch Q 5 (third switch) and the cathode of a diode D 6 , and the other end (source) of the switch Q 5 and the anode of the diode D 6 are connected to a joint of the secondary winding 5 b and the tertiary winding 5 c, and the cathode of the diode D 4 . The gate of the switch Q 5 is connected to one end (no circle side) of the tertiary winding 5 c via a waveform shaping circuit 11 serving as an ON-time setting unit.
The diode D 6 is connected between the drain and the source of the switch Q 5 . This diode D 6 is provided to suppress a spike voltage by absorbing energy of the spike voltage generated at recovery time of the diode D 4 in the capacitor C 5 . Specifically, a recovery current D 4 i flows counterclockwise through a path passing along D 3 , 5 b, D 4 , and D 3 at the recovery time of the diode D 4 , but by causing a current D 6 i to flow counterclockwise through a path passing along D 3 , 5 b, D 6 , C 5 , and D 3 , the spike voltage generated at the recovery time of the diode D 4 is suppressed.
The waveform shaping circuit 11 shapes the waveform of a voltage generated at the tertiary winding 5 c of the transformer T 2 , so as to set a drive voltage applied to the gate of the switch Q 5 being shorter than an ON-time of the switch Q 2 . FIG. 4 is an example of the waveform shaping circuit 11 . The waveform shaping circuit 11 has a time-constant that is determined according to a resistor RT, which having one end thereof connected to the tertiary winding 5 c and the other end thereof connected to the gate of the switch Q 5 , and a capacitor CT connected between the other end of the resistor RT and the ground.
The time-constant circuit having the resistor RT and the capacitor CT linearly increases the waveform of a voltage VCT across the capacitor CT in response to an input rectangular voltage waveform Vn 3 of the tertiary winding 5 c and applies this voltage VCT to the gate of the switch Q 5 , thereby the switch Q 5 is turned ON only when the voltage VCT falls within a range between a first threshold voltage VTH 1 and a second threshold voltage VTH 2 of the gate of the switch Q 5 .
›DESCRIPTION OF THE PREFERRED EMBODIMENTS · 2 of 5
The control circuit 10 a alternately turns ON/OFF the switches Q 1 and Q 2 in such a way that the width of an ON-pulse applied to the switch Q 1 is narrowed and the width of an ON-pulse applied to the switch Q 2 is widened as an output voltage of a load RL reaches or exceeds a reference voltage.
The control circuit 10 a turns the switch Q 1 ON state during a predetermined period of time after the switch Q 1 reaches a zero-voltage condition due to resonance of the capacitor C 1 connected to the switch Q 1 in parallel and a leakage inductance (not shown) of the primary winding and the secondary winding of the transformer T 2 .
Operations of the DC conversion apparatus according to the first embodiment will be explained next with reference to a timing chart under light load condition shown in FIG. 6 . In FIG. 6 , a drain-source voltage of the switch Q 1 is depicted by Q 1 v a drain current of the switch Q 1 is depicted by Q 5 i , a current flowing through the diode D 5 is depicted by D 5 i , a drain-source voltage of the switch Q 5 is depicted by Q 5 v , a drain current of the switch Q 5 is depicted by Q 5 i , a drain-source voltage of the switch Q 3 is depicted by Q 3 v , a gate voltage signal of the switch Q 1 is depicted by Q 1 g , a gate voltage signal of the switch Q 2 is depicted by Q 2 g , and a gate voltage signal of the switch Q 5 is depicted by Q 5 g , respectively.
When the switch Q 2 is turned OFF, resonance is produced by the capacitor C 1 and a leakage inductance existing between the primary and the secondary windings of the transformer T 2 , and then the voltage Q 1 v of the switch Q 1 decreases. And the voltage Q 1 v of the switch Q 1 becomes zero as the switch Q 1 is turned ON, thus zero-voltage switching of the switch Q 1 is achieved.
Next, when the switch Q 1 is turned ON, the current Q 1 i flows clockwise through a path passing along Vdc 1 , 5 a, Q 1 , and Vdc 1 . At this time, since a voltage is generated also at the secondary winding 5 b of the transformer T 2 , the switch Q 4 is turned ON and the current Q 4 i flows clockwise through a path passing along 5 b, Lo, Co, Q 4 , and 5 b, so that electric power is supplied to a load RL. At this time, Lo (Io) 2/2 of energy is stored in the smoothing reactor Lo, wherein Io is a current flowing into the smoothing reactor.
Subsequently, when the switch Q 1 is turned OFF, a current flows due to exciting energy stored at the primary winding 5 a, and then the capacitor C 1 is charged. At this time, resonance is produced by the capacitor C 1 and the leakage inductance between the primary and the secondary windings of the transformer T 2 , so that the voltage Q 1 v of the switch Q 1 increases.
When the electrical potential of the switch Q 1 reaches that of the clamp capacitor C 2 , the diode D 2 becomes conductive, and the clamp capacitor C 2 starts to be charged. At this time, zero-voltage switching of the switch Q 2 is achieved by turning ON the switch Q 2 . Furthermore, the voltage of the secondary winding 5 b of the transformer T 2 is reversed, and therefore the switch Q 4 is turned OFF and the switch Q 3 is turned ON. As a result, the current Q 3 i flows clockwise through a path passing along Lo, Co, Q 3 , and Lo due to the energy stored in the smoothing reactor Lo, so that the electric power is continuously supplied to the load RL. When the switch Q 2 is turned ON, the current Q 2 i flows through a path passing along C 2 , Q 2 , 5 a, and C 2 , and finally the energy is returned to the primary winding 5 a.
In a case where a load current decreases (under light load condition), the current of the smoothing reactor Lo flowing therethrough when the switch Q 1 is turned OFF (e.g., time t 2 ) becomes zero and the level is kept while the switch Q 1 is OFF, whereas the switch Q 3 remains ON state. Therefore, the electric charge stored in the smoothing capacitor Co is discharged, and then a current Q 3 i ′ flows through a path passing along Co, Lo, Q 3 , and Co, so that the energy is stored in the smoothing reactor Lo.
Next, when the switch Q 2 is turned OFF and the switch Q 1 is turned ON at time t 3 (same as time t 1 ), the switch Q 3 is turned OFF and the switch Q 4 is turned ON. Therefore, the energy stored in the smoothing reactor Lo is transferred to the capacitor C 5 via the diode D 5 . That is, the diode D 5 is activated and the energy is stored in the capacitor C 5 , so that a spike voltage is absorbed.
Subsequently, when the switch Q 1 is turned OFF and the switch Q 2 is turned ON at time t 4 (same as time t 2 ), a voltage at one end of the secondary winding 5 b of the transformer T 2 is applied to the gate of the switch Q 3 , thereby turning the switch Q 3 ON state.
The waveform shaping circuit 11 applies the gate voltage signal Q 5 g to the gate of the switch Q 5 to turn the switch Q 5 ON state within a period of time from time t 41 to time t 42 (same as time t 21 to time t 22 ) after the switch Q 2 is turned ON based on the voltage generated at one end of the tertiary winding 5 c of the transformer T 2 . Therefore, the energy in the capacitor C 5 is discharged through a path passing along C 5 , Q 5 , 5 b, D 3 , and C 5 . The diode D 3 is conductive because a current flows from the smoothing reactor Lo. Therefore, voltage is induced at the primary winding 5 a of the transformer T 2 .
The voltage induced at the primary winding 5 a causes a current Q 2 i ′ to flow clockwise through a path passing along 5 a, Q 2 (D 2 ), C 2 , and 5 a, and therefore, the clamp capacitor C 2 is charged. The energy stored in the clamp capacitor C 2 transfers to an exciting inductance of the transformer T 2 , and is returned to the DC power source Vdc 1 when the switch Q 1 is turned ON.
As shown in FIG. 6 , it is noted that the current Q 5 i flows during the ON-time of the switch Q 2 (while the gate voltage signal Q 2 g has H-level) and also during the ON-time T ON of the switch Q 5 (e.g., time t 21 to time t 22 or time t 41 to t 42 ), and that the energy is then returned to the clamp capacitor C 2 on the primary side of the transformer T 2 . The return amount of energy is determined by the duration of the ON-time of the switch Q 5 , so that the return amount of energy can be controlled by adjusting the ON-time of the switch Q 5 .
›DESCRIPTION OF THE PREFERRED EMBODIMENTS · 3 of 5
Since the voltage Q 3 v of the switch Q 3 is clamped, no spike voltage is generated, as shown in FIG. 6 . Therefore, it is possible to set a withstand voltage of the switch Q 3 to be a lower value. Accordingly, losses can be further reduced because a low ON-resistance device can be used.
As described above, according to the resonant DC conversion apparatus employing the active clamp topology in the first embodiment, the leakage inductance between the primary and the secondary windings of the transformer T 2 is increased, and the resonance between this leakage inductance and the capacitor C 1 is utilized so as to achieve the zero-voltage switching of the switches Q 1 and Q 2 . This leakage inductance makes the generation of the spike voltage more remarkable. However, this spike voltage can be eliminated by transferring the energy stored in the smoothing reactor Lo to the capacitor C 5 via the diode D 5 due to a current backflow caused under light load condition and returning this energy to the clamp capacitor C 2 on the primary side of the transformer T 2 without any loss by use of the switch Q 5 which is turned ON synchronously as the switch Q 2 is ON. Therefore, the withstand voltage of a rectification device can be lowered, and ON-resistance can be reduced by using a low withstand-voltage device. Furthermore, a C-R absorber for preventing the spike voltage can be eliminated, thereby simplifying the circuitry.
FIG. 7 is a signal timing chart of each component under heavy load condition of the DC conversion apparatus in a case where only the synchronous rectifying circuit is provided. FIG. 8 is a signal timing chart of each component under heavy load condition of the DC conversion apparatus according to the first embodiment. As shown in FIG. 7 , a spike voltage SP is generated across the switch Q 3 . In the first embodiment shown in FIG. 8 , the generated spike voltage is clamped through the diode D 5 by the voltage of the capacitor C 5 , so as to be suppressed. Furthermore, the diode D 6 is provided to absorb the energy of the spike voltage generated at recovery time of the diode D 4 in the capacitor C 5 , thereby suppressing the spike voltage. Therefore, low withstand voltage devices can be used for the switch Q 3 and the diode D 3 , leading to reducing losses. Similarly, low withstand voltage devices can be used for the switch Q 4 and the diode D 4 . Moreover, the waveform shaping circuit 11 can turn the switch Q 5 ON state within a range between the first threshold voltage VTH 1 and the second threshold voltage VTH 2 . That is, since the ON-time of the switch Q 5 is set to be shorter than that of the switch Q 2 , the current of the switch Q 5 under heavy load condition can be reduced. The ON-time of the switch Q 5 is adjustable by changing the second threshold voltage VTH 2 .
Second Embodiment
FIG. 9 is a circuit configuration diagram of a DC conversion apparatus according to a second embodiment. The DC conversion apparatus shown in FIG. 9 is different from that of the first embodiment shown in FIG. 3 only in that the waveform shaping circuit 11 is not provided. Other configurations shown in FIG. 9 are the same as those shown in FIG. 3 , so like reference signs denote like parts and details thereof will be omitted.
The DC conversion apparatus of the second embodiment also offers the same advantageous effect as that obtained from the DC conversion apparatus of the first embodiment.
Third Embodiment
FIG. 10 is a circuit configuration diagram of a DC conversion apparatus according to a third embodiment. The DC conversion apparatus shown in FIG. 10 is different from that of the first embodiment shown in FIG. 3 in that a transformer T 3 has a primary winding 5 a, a secondary winding 5 b, a tertiary winding 5 c, and a quaternary winding 5 d that is corresponding to an auxiliary winding of the present invention (the number of turns is n 4 ) and in that the primary winding 5 a is loosely coupled with the quaternary winding 5 d and is also tightly coupled with the secondary winding 5 b and the tertiary winding 5 c. Specifically, when the switch Q 1 is ON, the DC excitation of the transformer T 3 in its operating state is canceled by magnetomotive forces having the same magnitude but in opposite directions at the primary winding 5 a and the quaternary winding 5 d, and when the switch Q 1 is OFF, the DC excitation of the transformer T 3 in its operating state is canceled by magnetomotive forces having the same magnitude but in opposite directions at the secondary winding 5 b and the quaternary winding 5 d. Therefore, the exciting inductance can be increased, thereby reducing exciting current and thus reducing power losses.
The quaternary winding 5 d is provided at a position corresponding to the smoothing reactor Lo shown in FIG. 3 . The number of turns of the secondary winding 5 b of the transformer T 3 is the same as that of the quaternary winding 5 d of the transformer T 3 . The secondary winding 5 b and the tertiary winding 5 c of the transformer T 3 each have a winding phase opposite to that of the primary winding 5 a of the transformer T 3 , and the quaternary winding 5 d of the transformer T 3 has the same winding phase as that of the primary winding 5 a of the transformer T 3 .
The reverse mode voltage of the tertiary winding 5 c described herein means that the tertiary winding 5 c has a winding structure in relation to the primary winding 5 a in such that a switch Q 5 is turned OFF as the switch Q 1 is ON and that the switch Q 5 is turned ON as the switch Q 1 is OFF.
The anode of the diode D 5 is connected to a joint of the secondary winding 5 b and the cathode of the diode D 4 , one end (drain) of the switch Q 4 , and the gate of the switch Q 3 . The cathode of the diode D 5 is connected via the capacitor C 5 to the anode of the diode D 4 , the anode of the diode D 3 , and one end of the smoothing capacitor Co. A joint of the diode D 5 and the capacitor C 5 is connected to one end (drain) of the switch Q 5 and the cathode of the diode D 6 . The other end (source) of the switch Q 5 and the anode of the diode D 6 are connected to one end of the quaternary winding 5 d and the cathode of the diode D 3 . The gate of the switch Q 5 is connected to one end (no symbol side) of the tertiary winding 5 c via the waveform shaping circuit 11 .
›DESCRIPTION OF THE PREFERRED EMBODIMENTS · 4 of 5
Other configurations shown in FIG. 10 are the same as those in FIG. 3 , so like reference signs denote like parts and details thereof will be omitted.
FIG. 11 is a configuration diagram of the transformer provided in the DC conversion apparatus according to the third embodiment. The transformer shown in FIG. 11 includes a core 20 having a rectangular external shape and also having therein elongate apertures 24 a and 24 b that are formed to configure magnetic paths 25 a, 25 b, and 25 c in a longitudinal direction thereof. The primary winding 5 a, the secondary and the tertiary windings 5 b and 5 c closely and tightly coupled with the primary winding 5 a, and the quaternary winding 5 d loosely coupled with the primary winding 5 a are provided in a core part 20 a of the core 20 . In order to loosely couple the primary winding 5 a with the quaternary winding 5 d, a projecting part 20 b is formed between the secondary and the tertiary windings 5 b and 5 c, and the quaternary winding 5 d in the core part 20 a. A leakage flux increases at the projecting part 20 b, resulting in a large leakage inductance of the quaternary winding 5 d.
Operations of the DC conversion apparatus thus configured in the third embodiment will be explained with reference to a timing chart shown in FIG. 12 .
In FIG. 12 , a voltage across the switch Q 1 is depicted by Q 1 v a current flowing into the switch Q 1 is depicted by Q 1 i , a current flowing into the switch Q 2 is depicted by Q 2 i , a current flowing into the diode D 2 is depicted by D 2 i , and a current flowing into the quaternary winding 5 d of the transformer T 3 is depicted by n 4 i.
Operations performed under heavy load condition are explained first. When the switch Q 1 is turned ON at time t 1 , the current Q 1 i (equivalent to a current I 1 flowing through the primary winding 5 a ) flows clockwise through a path passing along Vdc 1 , 5 a, Q 2 , and Vdc 1 . At the same time, a voltage is generated at the quaternary winding 5 d that is loosely coupled with the primary winding 5 a, and then the current n 4 ia (equivalent to a current I 1 ′ corresponding to the current I 1 ) flows clockwise through a path passing along 5 d, Co, Q 3 , and 5 d, so that electric power is supplied to a load RL. The quaternary winding 5 d has a high leakage inductance as it is loosely coupled with the primary winding 5 a. At this time, the relationship of I 1 ·n 1 =I 1 ′·n 4 holds true according to the law of equal ampere-turns, and hence the DC excitation is canceled.
When the switch Q 1 is turned OFF at time t 2 , a current flows due to exciting energy stored in the primary winding 5 a, and then the capacitor C 1 is charged. At this time, voltage resonance is produced by a leakage inductance (not shown) of the primary winding 5 a of the transformer T 3 and the capacitor C 1 , and the voltage Q 1 v of the switch Q 1 increases abruptly.
When the electrical potential of the switch Q 1 reaches that of the clamp capacitor C 2 , the diode D 2 becomes conductive, and then the clamp capacitor C 2 starts to be charged as the current D 2 i (shown in FIG. 12 ) flows through the diode D 2 . At this time, zero-voltage switching of the switch Q 2 is achieved by turning the switch Q 2 ON state.
Subsequently, the clamp capacitor C 2 is completely charged, and the electrical charge stored therein is returned to the primary winding 5 a via the switch Q 2 . At this time, the filled circle side of the primary winding 5 a is negative, and the non-symbol side thereof is positive. Therefore, also on the secondary side of the transformer T 3 , the filled circle side of the secondary winding 5 b is negative and the non-symbol side thereof is positive, and the filled circle symbol side of the quaternary winding 5 d is negative and the non-symbol side thereof is positive. Furthermore, the current n 4 ib continues to flow clockwise through a path passing along 5 b, 5 d, Co, Q 4 , and 5 b due to a leakage inductance (not shown) of the quaternary winding 5 d. Therefore, a current always flows through the load RL irrespective of whether the switch Q 1 is ON or OFF, so that a ripple current of the smoothing capacitor Co can be reduced.
Since the secondary winding 5 b and the quaternary winding 5 d of the transformer T 3 have the same number of turns and opposite polarities, the magnetomotive forces of the both windings 5 b and 5 d are eliminated and become zero. That is, the DC excitation is canceled.
Accordingly, the impedance on the secondary side of the transformer T 3 seen looking from the primary side thereof is increased, and when the switch Q 2 is turned OFF at time t 3 (same as time t 1 ), the current of the switch Q 2 is almost enough to discharge the energy in the capacitor C 1 . Therefore, the electrical potential of the switch Q 1 decreases to zero, and then the diode D 1 becomes conductive. At this time, zero-voltage switching of the switch Q 1 can be achieved by turning ON the switch Q 1 .
On the other hand, in a case where a load current decreases (during light load condition), a current of the quaternary winding 5 d flowing therethrough when the switch Q 1 is turned OFF becomes zero while the switch Q 1 is OFF state whereas the switch Q 4 remains ON state. Therefore, the electric charge stored in the smoothing capacitor Co is discharged, and a current n 4 ib ′ flows through a path passing along Co, 5 d, 5 b, Q 4 , and Co, so that the energy is stored in the quaternary winding 5 d.
Next, the switch Q 2 is turned OFF and the switch Q 4 is also turned OFF. Therefore, the energy stored in the quaternary winding 5 d is transferred to the capacitor C 5 via the diode D 5 . That is, the diode D 5 is activated and the energy is stored in the capacitor C 5 , so that a spike voltage is absorbed.
Subsequently, the waveform shaping circuit 11 applies the gate voltage signal Q 5 g to the gate of the switch Q 5 thereby to turn ON the switch Q 5 after the switch Q 2 is turned ON (during the ON-time of the switch Q 2 ) according to the voltage generated at one end of the tertiary winding 5 c of the transformer T 3 . Therefore, the energy in the capacitor C 5 is discharged through a path passing along C 5 , Q 5 , 5 b, Q 4 , and C 5 (Q 54 i ), so that a voltage is induced in the primary winding 5 a of the transformer T 3 .
›DESCRIPTION OF THE PREFERRED EMBODIMENTS · 5 of 5
The voltage induced in the primary winding 5 a causes a current Q 2 i ′ to flow through a path passing along 5 a, Q 2 (D 2 ), C 2 , and 5 a, and therefore, the clamp capacitor C 2 is charged. The energy stored in the clamp capacitor C 2 transfers to an exciting inductance of the transformer T 3 , and is returned to the DC power source Vdc 1 as the switch Q 1 is turned ON. The return amount of energy is determined by the duration of the ON-time of the switch Q 5 , so that the return amount of energy can be controlled by adjusting the ON-time of the switch Q 5 .
Accordingly, since the voltage Q 4 v of the switch Q 4 is clamped, no spike voltage is generated. Therefore, it is possible to set a withstand voltage of the switch Q 4 to be a lower value. Therefore, losses can be further reduced because a low ON-resistance device can be used.
As described above, according to the DC conversion apparatus of the third embodiment, the energy stored in the quaternary winding 5 d due to a current backflow caused under light load condition is stored in the capacitor C 5 via the diode D 5 , and this energy is returned to the clamp capacitor C 2 on the primary side of the transformer T 3 without any loss by use of the switch Q 5 that is turned ON synchronously when the switch Q 2 is ON, so that the spike voltage can be eliminated.
The diode D 6 suppresses the spike voltage, by absorbing the energy of the spike voltage generated at recovery time of the diode D 3 in the capacitor C 5 . Therefore, a low withstand voltage device can be used, thereby reducing power losses.
The quaternary winding 5 d is provided on the secondary side of the transformer T 3 , the primary winding 5 a and the quaternary winding 5 d are loosely coupled with each other, and the primary winding 5 a and the secondary winding 5 b are tightly coupled with each other. When the switch Q 1 is ON, the DC excitation of the transformer T 3 in its operating state is canceled by magnetomotive forces having the same magnitude but in opposite directions at the primary winding 5 a and the quaternary winding 5 d, and when the switch Q 1 is OFF, the DC excitation of the transformer T 3 in its operating state is canceled by magnetomotive forces having the same magnitude but in opposite directions at the secondary winding 5 b and the quaternary winding 5 d. Therefore, the exciting inductance can be increased, leading to a reduced exciting current and thus to reduced losses. Furthermore, zero-voltage switching can be achieved, and a compact, highly efficient, and low-noise DC conversion apparatus can be obtained.
Fourth Embodiment
FIG. 13 is a circuit configuration diagram of a DC conversion apparatus according to a fourth embodiment. The DC conversion apparatus shown in FIG. 13 is different from that of the third embodiment shown in FIG. 10 only in that the waveform shaping circuit 11 is removed. Other configurations shown in FIG. 13 are the same as those shown in FIG. 10 , so like reference signs denote like parts and details thereof will be omitted.
The DC conversion apparatus in the fourth embodiment can also offer the same advantageous effect as that obtained from the DC conversion apparatus in the third embodiment.
›Advantages of the Invention
According to the present invention, the energy stored in a smoothing reactor due to a current backflow caused under light load condition is stored in a capacitor via a diode, and this energy is returned to a clamp capacitor on the primary side without any loss by use of a third switch that is turned ON synchronously when a second switch is ON, so that a spike voltage can be eliminated. Therefore, a withstand voltage of a device used for a synchronous rectifier can be lowered, so that a low withstand-voltage device can be used. Furthermore, a circuit that causes losses, such as a surge absorber, is not necessary, and therefore losses can be reduced. Accordingly, a highly efficient and compact DC conversion apparatus can be obtained.
›INDUSTRIAL APPLICABILITY
The present invention is applicable to a switching power supply, such as a DC-DC converter and an AC-DC converter.
This application claims benefit of priority under 35USC §119 to Japanese Patent Applications No. 2004-381758, filed on Dec. 28, 2004, the entire contents of which are incorporated by reference herein. Although the invention has been described above by reference to certain embodiments of the invention, the invention is not limited to the embodiments described above. Modifications and variations of the embodiments described above will occur to those skilled in the art, in light of the teachings. The scope of the invention is defined with reference to the following claims.
Claims as granted
12 claimsLog in to read the claims of this application.
Log in to unlockClassifications
4 codes- H02M3/335
Claim changes
SoonSee which claims were amended, added or cancelled during examination, with every added and removed word marked.
The published claims of this application are not paired with the granted ones in what we hold.
File wrapper
See the full prosecution history — every USPTO and applicant action on this file, in order.
Log in to unlockDocuments
Log in to open the documents of this file: the application as filed, every office action and response, the notice of allowance.
Log in to unlockChain of title
See the full assignment history — every owner this patent has passed through, with recordation dates and reel/frame numbers.
Log in to unlock