JP6424733B2 - Power converter controller - Google Patents

Description

  The present invention relates to a power converter control device that alternately generates positive and negative voltage pulses and applies them to a transformer of a power converter.

  2. Description of the Related Art Conventionally, push-pull, full-bridge, and other insulated DCDC converters (power converters) are known which alternately apply positive and negative voltage pulses to a transformer to convert voltage. In such a power converter, if the voltage pulse is forcibly turned off midway due to over current detection or the like, biased magnetism occurs due to the unbalance of the positive and negative applied voltages. When the transformer reaches magnetic saturation due to biased magnetization, there is a problem that a large current flows in the device. For this problem, for example, Patent Document 1 discloses a technique for adjusting the width of the pulse following the forcibly turned off pulse to suppress instantaneous magnetic deviation.

JP, 2010-161843, A

The prior art of Patent Document 1 is considered to be effective when two switching elements that alternately generate positive and negative voltage pulses perform switching operation continuously. However, it is not possible to cope with the case where the switching operation is completely stopped after the switching operation has completely stopped for one pulse, that is, a half cycle or more at the time of the occurrence of the overcurrent or the output sudden change. Also, there is no technical idea of adjusting the pulse width prior to the pause of a period of half a cycle or more.
The present invention has been made in view of such a point, and an object thereof is to provide a power converter control device for suppressing the bias magnetism of the transformer when the switching operation is completely stopped for a half cycle or more. It is to do.

  A power converter to which the power converter control device of the present invention is applied includes a transformer for converting power between an input end connected to a power supply and an output end connected to a load, and positive and negative voltage pulses to the transformer. A plurality of switching elements that perform switching operation so as to apply alternately, a rectifying element connected to an output path of a transformer, and a reactor connected between the rectifying element and an output end. As this power converter, for example, an isolated DCDC converter of a push-pull type, a full bridge type or the like corresponds.

A power converter control device according to the present invention, which is applied to the above power converter and controls energization of a transformer, has a pulse width setting unit and a gate pulse signal generation unit.
The pulse width setting unit sets an “operation switch” which is a switching element to be turned on next and a “pulse width” which is an on time of the operation switch.
The gate pulse signal generation unit generates a gate pulse signal for operating the plurality of switching elements based on the operation switch and the pulse width set by the pulse width setting unit.

The pulse width setting unit is applied to the transformer immediately before the “sleep period” in which the switching operation is paused for a half cycle or more of the voltage pulse (in the reference mode, at the time of rest when the pause period ends and restarts the switching operation ) . The operation switch and the pulse width are set to balance positive and negative voltages.

The power converter control device according to the first aspect is characterized by the operation of the pulse width setting unit immediately before the idle period. When the pulse width setting unit according to the first aspect externally receives a stop signal instructing to stop the current switching operation and shift to the pause period, the magnetic flux generated in the transformer by application of the voltage pulse is made zero. The pulse width of the last pulse, which is the last voltage pulse before the pause period, is set to shift to the pause period in the close state.
Specifically, the power converter control device receives a stop signal instructing to stop the current switching operation and shift to the pause period at a timing after an intermediate time in the on period of either the positive or negative voltage pulse. Simultaneously turns off the current voltage pulse simultaneously with the reception of the stop signal. The pulse width setting unit is a voltage pulse of the next opposite polarity so as to shift to the pause period with the magnetic flux generated in the transformer being brought close to zero by the application of the voltage pulse, and the last voltage pulse before the pause period. The pulse width of a certain last pulse is set to a pulse width corresponding to the time when the current voltage pulse is turned on past the middle time.

  The first aspect presupposes a configuration in which the switching element performs switching operation one more time after receiving the stop signal and can generate the last pulse. In this aspect, when the stop signal is received, by adjusting the pulse width of the last pulse, it is possible to suppress the bias magnetism of the transformer at the time of transition to the idle period.

The power converter control device according to the second aspect, which is the reference aspect, is characterized in the operation of the pulse width setting unit at the time of recovery after the idle period. When the pulse width setting unit according to the second aspect receives from the outside a return signal instructing termination of the idle period and resuming the switching operation from the outside, the last voltage pulse before the idle period in the return first switching operation The operation switch is set to apply to the transformer a voltage pulse of the opposite polarity to the last pulse.
When a voltage pulse of the same polarity as the last pulse is applied for the first time of recovery, the imbalance between the positive and negative applied voltages increases, and the transformer may reach magnetic saturation. Therefore, magnetic saturation can be prevented by applying a voltage pulse of the opposite polarity to the last pulse at the first recovery.

In the pulse width setting unit according to the second aspect, preferably, for the excitation current and the magnetic flux generated in the transformer by application of the voltage pulse during the idle period, the positive and negative of the magnetic flux generated by the reduction of the absolute value of the excitation current during the idle period. Specifically, based on the value of the ET product at the time of recovery or the value of the excitation current, the pulse width of the recovery first time is set so as to compensate for the unbalance of In this aspect, when the return signal is received after the idle period, it is possible to suppress the biased magnetism of the transformer at the return time point by adjusting the pulse width of the return first time.
Furthermore, the pulse width setting unit may determine that the ET product is reset when the value of the ET product remaining at the time of recovery is equal to or less than a predetermined value, and may set the pulse width at the start of steady operation.

BRIEF DESCRIPTION OF THE DRAWINGS The schematic block diagram of the push pull type | mold insulation DCDC converter to which the power converter control apparatus by embodiment of this invention is applied. The time chart of the pulse signal generated by a 1st embodiment of the present invention. The time chart of the pulse signal generated by a 2nd embodiment of the present invention. The time chart of the pulse signal generated by a 3rd embodiment of the present invention. The time chart of the pulse signal generated by a 4th embodiment of the present invention. The time chart explaining ET product reset decision. The figure which shows the example of a sensor arrangement | positioning for ET product reset determination. The figure explaining calculation of the correction | amendment pulse width of the restoration first time. The figure explaining the voltage applied to a transformer secondary side during operation stop. The figure which shows the change of the remaining ET product with the time from operation stop. The time chart of the pulse signal generated by a 5th embodiment of the present invention. The time chart of the pulse signal by the power converter control apparatus of a comparative example. BRIEF DESCRIPTION OF THE DRAWINGS The schematic block diagram of the full bridge type isolated DCDC converter to which the power converter control apparatus by embodiment of this invention is applied.

Hereinafter, power converter control devices according to a plurality of embodiments of the present invention will be described based on the drawings. In a plurality of embodiments, substantially the same configuration is given the same reference numeral and the description is omitted. The second embodiment corresponds to a mode for carrying out the invention according to the claims.
This power converter control device is applied to an isolated DCDC converter (power converter) such as a push-pull type or a full bridge type that alternately applies positive and negative voltage pulses to a transformer to convert the voltage, and controls energization of the transformer Device.

[Configuration of power converter]
First, as an example of a power converter to which this power converter control device is applied, a schematic configuration of a push-pull type isolated DCDC converter will be described with reference to FIG.
The push-pull converter 101 is provided between the input end 11 connected to a power supply such as a battery and the output end 12 connected to a load such as a motor, and the transformer 201, switching elements SW1 and SW2, diodes (rectification Elements 51, 52, reactor 7, smoothing capacitors C1, C2, etc. are provided.

  The transformer 201 converts power between the primary side connected to the input end 11 and the secondary side connected to the output end 12. The transformer 201 includes a first transformer unit 21 composed of a primary coil 31 and a secondary coil 33, and a second transformer unit 22 composed of a primary coil 32 and a secondary coil 34. One ends of the primary coils 31 and 32 are connected to each other at the primary connection point J1, and one ends of the secondary coils 33 and 34 are connected to each other at the secondary connection point J2.

The primary side connection point J 1 is connected to one first terminal 111 of the input terminal 11. The other ends of the primary coils 31 and 32 are connected to the second terminal 112 of the input end 11 via the switching elements SW1 and SW2, respectively. Further, a primary smoothing capacitor C1 is connected between the two terminals 111 and 112 of the input terminal 11.
The switching elements SW1 and SW2 perform switching operation so that positive and negative voltage pulses are alternately applied to the primary coils 31 and 32 of the transformer 201. In the following description of the embodiment, the “switching element” is appropriately omitted and simply referred to as “SW1” or “SW2”.

In the present embodiment, MOSFETs (metal oxide semiconductor field effect transistors) having parasitic diodes are used as SW1 and SW2. The drains of SW <b> 1 and SW <b> 2 are connected to the primary coils 31 and 32, and the sources are connected to the second terminal 112 of the input terminal 11. A gate pulse signal is input from the gate pulse signal generation unit 82 of the control device 80 to the gates of SW1 and SW2.
In another embodiment, a transistor such as an IGBT may be used as the switching element. Also, a reflux diode may be used instead of the parasitic diode.

  The secondary side connection point J 2 is connected to the first terminal 121 of the output end 12. The other ends of the secondary coils 33 and 34 are connected to the second terminal 122 of the output end 12 via the diodes 51 and 52 and the reactor 7, respectively. The diodes 51 and 52 are connected to the output path on the secondary side of the transformer 201 to rectify the secondary current. The reactor 7 stores magnetic energy by energization. In addition, a secondary smoothing capacitor C2 is connected between the two terminals 121 and 122 of the output terminal 12.

The power converter control device (hereinafter simply referred to as “control device”) 80 has a pulse width setting unit 81 and a gate pulse signal generation unit 82.
The pulse width setting unit 81 sets an “operation switch”, which is a switching element to be turned on next, and a “pulse width”, which is the on time of the operation switch. That is, the pulse width setting unit 81 sets which switching element is turned on for how long.

Specifically, the pulse width setting unit 81 acquires, as the power information on the secondary side of the transformer 201, the reactor current flowing through the reactor 7 and the output voltage between the electrodes of the secondary smoothing capacitor C2, and the voltage and current based on those information. The ratio (conduction ratio) of the on time of SW1 and SW2 to the switching cycle is calculated by control system calculation. Then, basically, SW1 and SW2 are alternately set as operation switches. Also, based on the result of the voltage / current control system calculation, the pulse widths of SW1 and SW2 are set to the same length.
The gate pulse signal generation unit 82 generates a gate pulse signal based on the operation switch and the pulse width set by the pulse width setting unit 81, and outputs the gate pulse signal to the gates of SW1 and SW2. Thereby, SW1 and SW2 perform switching operation alternately for the same time.

Voltage pulses with opposite polarities are applied to the transformer 201 when SW1 is turned on and when SW2 is turned on. In this embodiment, a voltage applied when SW1 is turned on is defined as positive, and a voltage applied when SW2 is turned on is defined as negative.
Thus, the control device 80 operates SW1 and SW2 so that positive and negative voltage pulses are alternately applied to the transformer 201 during steady operation. A period in which SW1 and SW2 are turned on once and positive and negative voltage pulses are applied is referred to as "one cycle" of the switching operation.

Here, in the case where the switching operation is paused for one pulse, that is, a half cycle or more, the period is referred to as a "pause period". Before and after the idle period, biased magnetism may occur due to an imbalance between positive and negative applied voltages. When the transformer 201 reaches magnetic saturation due to biased magnetization, a large current flows in the converter 101.
In addition, in patent document 1 (Unexamined-Japanese-Patent No. 2010-161843), when switching operation is interrupted momentarily in a short period less than a half cycle, the technique of adjusting the width of the pulse after the instantaneous interruption and suppressing a biased magnetism It is disclosed. However, no mention is made of the suppression of the bias magnetization when the switching operation is stopped for a half cycle or more.

Therefore, the control device 80 according to each embodiment of the present invention is characterized in that the switching operation is adjusted before and after the idle period over a period of a half cycle or more to suppress the bias magnetism. Therefore, the pulse width setting unit 81 “stops the current switching operation and instructs stop signal Q to shift to the idle period”, and “return to instruct the termination of the idle period to restart the switching operation. (Restart) signal “RS” is received.
The reason why the stop signal Q is commanded, that is, the reason why the pause period occurs is not limited. For example, it may be due to an unintended sudden phenomenon such as overcurrent detection described in Patent Document 1, or may be a predetermined reason.

  Hereinafter, a specific adjustment method of the switching operation performed by the control device 80 before and after the idle period will be described in detail for each embodiment. In the first and second embodiments, an example of adjusting the voltage pulse immediately before the pause period will be described. Further, in the third to fifth embodiments, an example will be described in which the voltage pulse at the time of recovery after the idle period ends and the switching operation is resumed.

[Adjustment of voltage pulse]
First and second embodiments
The first and second embodiments of the present invention will be described with reference to FIGS. 2 and 3. The first and second embodiments are characterized by the operation of the pulse width setting unit 81 immediately before the pause period.
The timing charts of FIG. 2 and FIG. 3 show the switching operation of SW1 and SW2 before and after the idle period, the pulse of the transformer applied voltage generated thereby, the timing of the transformer excitation current Im, and the stop signal Q. When SW1 is turned on, a pulse of positive voltage is applied, and the transformer excitation current Im increases. When SW2 is turned on, a pulse of negative voltage is applied, and the transformer excitation current Im decreases. When both SW1 and SW2 are off, no voltage pulse is generated, and the transformer excitation current Im hardly changes.

For example, from time t90 to time t92, SW2 turns on for a period of a pulse width To set by the voltage / current control system calculation. The operation at this time is called "stationary operation".
Throughout the following embodiments, it is assumed that the power supply and load conditions do not change significantly before and after the idle period. Therefore, it is assumed that the pulse width To set by the voltage current control system calculation is constant before and after the idle period.

Further, the product (E × T _ON ) of the transformer applied voltage and the on time of the switching element is referred to as an “ET product”. In the steady operation, the ET product (E × To) by the on operation of SW1 and the ET product (−E × To) by the on operation of SW2 are equal in absolute value and opposite in sign. While the SW1 and SW2 are being repeatedly switched, the ET product balance is maintained. As described later, the ET product value reflects the change in magnetic flux.

  In the control device 80 and the converter 101 in the first and second embodiments, after the pulse width setting unit 81 receives the stop signal Q during the steady operation, the SW1 or SW2 performs the switching operation one more time, and the last before the idle period It is assumed that the "last pulse" which is a voltage pulse of can be generated. When receiving the stop signal Q, the pulse width setting unit 81 sets the pulse width of the last pulse such that the magnetic flux generated in the transformer 201 is brought close to zero by application of a voltage pulse to shift to the idle period.

  The fact that the magnetic flux generated in the transformer 201 is substantially zero is referred to as "the magnetic flux is reset". FIGS. 2 and 3 show the transition to the quiescent period in a state where the excitation current Im is ideally zero, that is, in a state where the magnetic flux is completely reset. In the example of FIG. 2, the pulse width of the last pulse due to the on operation of SW1 from time t94 is adjusted so that the magnetic flux is reset. Further, in the example of FIG. 3, the pulse width of the last pulse due to the on operation of SW2 from time t98 is adjusted.

  However, the excitation current Im at the transition to the quiescent period may not be exactly zero, that is, the magnetic flux may not be completely reset. In short, it is the technical idea of the first and second embodiments that the pulse width of the last pulse is adjusted in the direction in which the magnetic flux approaches zero as compared to the state immediately before the last pulse. Thereby, it is possible to suppress the bias magnetism of the transformer 201 at the time of transition to the idle period.

The timing of receiving the stop signal Q differs between the first embodiment and the second embodiment. Here, the time when half of the pulse width To has elapsed from the start of the pulse width To is referred to as "intermediate time". If the positive and negative voltage pulses in the steady operation are symmetrical, the transformer excitation current Im passes through the zero point in the middle.
In the first embodiment shown in FIG. 2, "a timing (solid line arrow) during a period (time t92 to t94) during which both positive and negative voltage pulses are off" or "a voltage pulse of either positive or negative (SW1 in this example) The stop signal Q is received in the timing (broken line arrow) before the intermediate time t95 in the on period (time t94 to) of the positive voltage pulse).

In this case, the pulse width setting unit 81 sets the pulse width of the last pulse due to the on operation of SW1 to a half (To / 2) of the pulse width To in the steady operation. Therefore, the application of the positive voltage pulse ends at the middle time t95. As a result, it is possible to shift to the quiescent period with the excitation current Im being exactly zero.
In addition, at the first return after the idle period, “a pulse width at the start of steady operation” is set on the premise that the excitation current Im starts from zero. That is, from time t11 to time t12 corresponding to the middle time, for example, the pulse width of SW2 is set to (To / 2). Thereafter, the switching operation is repeated such that the switch SW1 is turned on at times t14 to t16, the switch SW2 is turned on at times t20 to t22, and so on with the pulse width To in the steady operation.

In the second embodiment shown in FIG. 3, “a timing t95 q (solid line arrow) after an intermediate time t 95 in an on period (time t 94-) of either positive or negative voltage pulse (a positive voltage pulse by SW 1 in this example)” A stop signal Q is received. At this time, the current voltage pulse is turned off at the same time as the stop signal Q is received. As a result, the pulse width Tinc of the incomplete voltage pulse turned off halfway is longer than half of the pulse width To in steady operation as shown in equation (1.1), and the pulse in steady operation is It is shorter than the width To.
(To / 2) <Tinc <To (1.1)
Further, when the difference between Tinc and (To / 2) is "excess time Tex", Tinc is expressed by equation (1.2).
Tinc = (To / 2) + Tex (1.2)

  In this case, the pulse width setting unit 81 sets the pulse width of the next voltage pulse of the opposite polarity (in this example, a negative voltage pulse by SW2) to the current voltage pulse (positive voltage pulse by SW1) after the middle time t95. Set the pulse width corresponding to the on time Tex. Thus, a voltage pulse of the opposite polarity (negative voltage pulse by SW2) is applied as the last pulse from time t98 to t99. In short, the positive voltage pulse applied to the excess time Tex from time t95 to t95 q is canceled by the negative voltage pulse at the same time Tex from time t98 to t99. As a result, it is possible to shift to the quiescent period with the excitation current Im being exactly zero.

Third Embodiment
Next, the third to fifth embodiments of the present invention are characterized by the operation of the pulse width setting unit 81 at the time of return after the idle period. First, the third embodiment will be described with reference to FIG.
In FIG. 4, the control device 80 turns on SW2 with pulse width To during steady operation from time t90 to t92, turns on SW1 with pulse width To during steady operation from time t94 to t96, and then transitions to the idle period. doing. First, as in the second embodiment, since the rest period just before not adjust the pulse width of the last pulse, the transformer magnetizing current Im ₀ start time t96 of the quiet period, non-zero value (in the example of FIG. 4 positive The value of

  When receiving the recovery signal RS during the idle period, the pulse width setting unit 81 applies a voltage pulse of the opposite polarity to the last pulse, which is the last voltage pulse before the idle period, to the transformer 201 in the switching operation of the recovery first time. Set the operation switch to In the example of FIG. 4, since the polarity of the last pulse is a positive voltage by SW1, SW2 is set as an operation switch so that a negative voltage is applied at the first recovery.

  If the switching operation continues without pausing, the time when SW2 is scheduled to turn on based on the clock cycle is set to t10. Then, when the return signal RS is input immediately before time t10 (during the time t06 to t10) (solid arrow), it is sufficient to turn on the operation switch SW2 at time t10 immediately after. On the other hand, when the return signal RS is input immediately before time t04 when the SW1 is scheduled to turn on based on the clock cycle (dotted line arrow), as shown by the two-dot chain line After waiting for time t10, the operation switch SW2 is turned on.

  Here, FIG. 12 is referred for the behavior of the comparative example. This comparative example does not have the function of determining the first return operation switch after the idle period based on the polarity of the last pulse. Therefore, when the return signal RS is received, one of SW1 and SW2 turns on according to the situation. In the example shown in FIG. 12, when the polarity of the last pulse is a positive voltage by SW1, the voltage pulse at the first recovery is also a positive voltage by SW1. That is, voltage pulses of the same polarity are applied to the transformer 201 twice in succession, with the idle period interposed therebetween.

At this time, when returning after the idle period, the positive excitation current Im is further superimposed on the basis of the transformer excitation current Im ₀ (> 0) immediately before the idle period. In order to prevent the magnetic saturation of the transformer 201 in this situation, the upper limit value of “2 UL” which is about twice the upper limit value UL of the excitation current Im (or the ET product) assumed in steady operation is set as the upper limit value. , Need to design a transformer specification. Therefore, the size of the transformer 201 and the wiring portion can not be increased.

  On the other hand, in the third embodiment, the operation switch is set such that a voltage pulse of the opposite polarity to the last pulse is applied to the transformer 201 in the switching operation of the first return. As a result, voltage pulses of the same polarity are continuously applied to the transformer 201 twice across the idle period, and overlapping of the excitation current Im of the same polarity is avoided. Therefore, since the transformer specification may be designed based on the upper limit value UL of the excitation current Im (or the ET product) assumed in the steady operation, the transformer 201 and the wiring portion can be miniaturized.

  By the way, under the condition where the microcomputer 84 constituting the control device 80 operates stably, information such as the polarity and timing of the last pulse is stored in the RAM at the start of the pause period, and the RAM is accessed at the time of recovery. Read out. However, if the control power is interrupted or the microcomputer is reset during the idle period, the storage of the RAM may be erased.

Therefore, as shown in FIG. 4, the control device 80 has a storage device 85 outside the microcomputer 84. At the start of the pause period, the microcomputer 84 stores information on the polarity and timing of the last pulse in a storage device 85 such as a flash memory (EEPROM). The microcomputer 84 can read the information of the last pulse from the storage device 85 even when no storage remains in the RAM when returning after the pause period. Thus, the pulse width setting unit 81 can set the operation switch and the pulse width based on the read information.
In the third embodiment, the operation switch for the first return is set mainly based on the polarity information of the last pulse. On the other hand, when the storage device 85 is applied to the next fourth embodiment, the pulse width of the recovery first time is set based on the timing information of the last pulse.

Fourth Embodiment
A fourth embodiment of the present invention will be described with reference to FIGS. The fourth embodiment shares the features of the third embodiment regarding the setting of the operation switch for the first return. Furthermore, the pulse width setting unit 81 is an imbalance between positive and negative of the magnetic flux generated by the decrease of the absolute value of the excitation current Im during the idle period, with respect to the excitation current Im and the magnetic flux generated in the transformer 201 by application of the voltage pulse before the idle period. Set the recovery first pulse width to compensate for.
As shown in FIG. 5, the absolute value Im ₀ rest period initial exciting current gradually decreases with the operation stop signal time T _Q, the return time t10 is reduced to Imr.

First, the ET product reset determination will be described with reference to the time chart of FIG. The fact that the value of the ET product is substantially zero is referred to as "the ET product is reset." FIG. 6 shows changes in the secondary side reactor current IL, the voltage (DS) (drain-source) of the switching elements SW1 and SW2, and the ET product before and after stopping the switching operation. The voltage between DS of SW1 and SW2 is in a relation in which the positive and negative are inverted.
During steady-state operation, reactor current IL pulsates centering on steady-state value IL_const with switching operation of SW1 and SW2. Also, the ET product alternates between positive and negative values.

  After stopping the switching operation at time t0, the value of the ET product maintains a substantially constant value in period I up to time t1, and has not been reset yet. In period II from time t1 to time t2, the ET product is being reset. Then, in the period III after time t2, the ET product is reset.

More specifically, in period I, there is a relationship of “reactor current IL> excitation current Im”. The exciting current Im circulates on the secondary side, and the magnetic flux of the transformer 201 is substantially constant.
In period II, the relationship “reactor current IL <excitation current Im” is established. The exciting current Im, which is larger than the reactor current IL, flows on the primary side. As a result, a voltage is applied to the transformer 201, and the value of the ET product, that is, the magnetic flux changes. Moreover, the voltage between DS of SW1 and SW2 resonates.
In period III, reactor current IL becomes zero. Moreover, the resonance amplitude of the voltage between DS of SW1 and SW2 is suppressed.

It is determined that the ET product has been reset in the following cases.
(1) When a predetermined time has elapsed after detection of ID1 = 0 or ID2 = 0 for diode currents ID1 and ID2 flowing through diodes 51 and 52 on the secondary side (2) predetermined after detection of reactor current IL = 0 When time has elapsed (3) When the voltage between DS of SW1 and SW2 starts resonance for a predetermined time or when a predetermined time elapses after the resonance amplitude is suppressed (4) excitation current Im = 0 When it detects

Here, FIG. 7 will be referred to for an exemplary arrangement of sensors for determining ET product reset. The numbers (1) to (4) correspond to the numbers of the above-mentioned reset determination method.
(1), (2) The diode currents ID1 and ID2 and the reactor current IL detect, for example, the voltages across the shunt resistors 65, 66, and 67 connected in series to the respective elements. The detection voltage may be input to the control device 80 via the isolator 86 as needed. Also, a Hall element may be used instead of the shunt resistor.

(3) The voltage (Vtr1, Vtr2) between DS of SW1 and SW2 detects the voltage division by the resistors 63 and 64 connected in parallel to one of the switching elements (for example, SW2).
(4) The excitation current Im is detected by passing the four wires of the primary coils 31 and 32 and the secondary coils 33 and 34 to the clamp type current sensor 60.

As described above, the pulse width setting unit 81 performs the ET product reset determination when returning.
When the ET product is reset, the pulse width setting unit 81 sets the pulse width at the start of the steady operation and restores the switching operation. That is, similarly to FIGS. 2 and 3 of the first and second embodiments, the pulse width of the first return is set to half (To / 2) of the pulse width of the steady operation.
On the other hand, when the ET product is not reset, the pulse width setting unit 81 calculates and sets the correction pulse width T ^* as the pulse width for the first recovery. Subsequently, a method of calculating the correction pulse width T ^* will be described.

In as indicated by the two-dot chain line 5, it is assumed that the exciting current Im ₀ during the suspension period unchanged. At this time, in the first return, a voltage pulse having a pulse width To at the time of steady operation is generated at time t10 to t12 by the on operation of SW1. Therefore, when the absolute value of the exciting current by a time Tr corresponding to the amount that has been reduced from Im ₀ to Imr, delaying the on-timing of SW1 from the time t10 to the time t10r generating correction pulse width T ^* voltage pulse, SW1 is The excitation current at time t12 when it is turned off can be made the same level as in the steady operation.

  Thus, due to the decrease in the absolute value of the excitation current Im during the idle period, positive and negative imbalance of the magnetic flux occurs at the first return. For example, after the change to the zero point (positive side in the example of FIG. 5) of the change of the excitation current Im due to the voltage pulse of the first recovery, α is over the zero point (negative side in the example of FIG. 5) If the change is β, then α <β.

FIG. 8 is an enlarged view of the change of the excitation current Im at the time of return of FIG.
Assuming that the times corresponding to the changes α and β of the excitation current Im are Tα and Tβ, the correction pulse width T ^* is represented by the sum of the times Tα and Tβ. Further, Tβ corresponds to one half (To / 2) of the pulse width To set by the voltage current control system calculation. Therefore, the pulse width setting unit 81 calculates the time Tα as follows, and calculates the correction pulse width T ^* of the first recovery.

Refer to FIG. 9 for the voltage on the secondary side of the transformer 201 during the stop of operation.
The on-voltage of the diode 51 and 52 Vf1, Vf2, the resistance component of the transformer 201 and the wiring R, the value of the residual ET product in operation stop time T _Q and ET _Q. Further, the excitation current Im and the value ET _Q of the residual ET product can be expressed as a function of time t after the switching operation is stopped, as expressed by equation (2).
ET _Q (t) = (Vf1 + Vf2 + Im (t) × R) × T _Q (2)
That is, the on voltages Vf1 and Vf2 of the diodes 51 and 52 and the drop voltage due to the resistor R are applied to the transformer 201.

The time change of the residual ET product value ET _Q is represented by the characteristic shown in FIG. Controller 80, for example with reference to the characteristic map, it is possible to estimate the value ET _Q residual ET product in operation stop time T _Q. Then, based on the value ET _Q residual ET product during recovery, it is possible to use the equation (3), calculates the correction pulse width T ^* of a return first.
T ^* = Tα + Tβ
= ET _Q / E + (To / 2) (3)
When there is a time difference between the time of calculation and the time of return, the time difference may be corrected.

Further, another calculation method of the time Tα utilizes a relation (Equation (4.1)) that the product of the excitation inductance Lm and the excitation current Im is an ET product. In this method, based on the value Imr of the excitation current at the time of recovery, the correction pulse width T ^* of the recovery first time can be calculated using Equation (4.2).
ET = Lm × Im (4.1)
T ^* = Tα + Tβ
= (Lm x Imr) / E + (To / 2) (4.2)

As described above, in the fourth embodiment, when the ET product is not reset at the time of recovery after the idle period, the recovery first time based on the value ET _Q of the remaining ET product at recovery time or the value Imr of the excitation current The correction pulse width T ^{* of} is calculated. Thereby, it is possible to more appropriately suppress the bias magnetism of the transformer 201 at the time of return.
In addition, when the ET product has already been reset, it is not necessary to suppress the biased magnetization, so the pulse width at the start of the steady operation is set and the switching operation is restored. Similar results can be obtained even if ET _Q 0 0 and Imr 0 0 in the equations (3) and (4.2). Thereby, the calculation of the control device 80 can be simplified.

Fifth Embodiment
A fifth embodiment of the present invention will be described with reference to FIG. The fifth embodiment shares the features of the third embodiment regarding the setting of the operation switch for the first return. Furthermore, the pulse width setting unit 81 is set to gradually increase a plurality of pulse widths after the first recovery as “soft start” at the time of recovery after the idle period.

In the example shown in FIG. 11, the pulse width T2a of time t10 to t12 by the ON operation of SW2 for the first recovery, the pulse width T1a of t14 to t16 by the ON operation of SW1 for the second recovery, and the ON operation of SW2 for the third recovery The pulse width T2b of the time t20 to t22 is set to be equal to or less than the pulse width To at the time of steady operation and to be sequentially longer.
However, “a plurality of times” at which soft start is performed means at least two times, and at least the first return and the second return, that is, one cycle of the pulse width may be set shorter than the pulse width To in steady operation. Thereby, the driving of the load can be started gently.

  The fifth embodiment differs from the fourth embodiment in that it does not completely eliminate positive and negative voltage imbalances with a single return first pulse, and some imbalances occur. However, as in the comparative example (see FIG. 12), an imbalance that requires twice the excitation current or the ET product safety factor does not occur. Therefore, it is possible to appropriately suppress the biased magnetization of the transformer 201 while realizing the soft start.

(Other embodiments)
As another example of a power converter to which the power converter control device of the present invention is applied, FIG. 13 shows a full bridge type isolated DCDC converter. In the full bridge converter 102, on the input side of the transformer 202, a full bridge circuit is configured by the switching elements SW1 to SW4 of the two pairs of upper and lower arms. The switching elements SW1 and SW4 on one diagonal of the full bridge circuit and the switching elements SW2 and SW3 on the other diagonal are alternately turned on and off, and positive and negative voltages Vtr are alternately applied to the primary coil 35 of the transformer 202

The gate pulse signal generation unit 82 of the control device 80 outputs a gate pulse signal with SW1 and SW4 as one pair and SW2 and SW3 as the other pair. The other features are the same as in the case of the push-pull converter 101.
As mentioned above, the present invention is not limited to the above-mentioned embodiment at all, and can be implemented in various forms in the range which does not deviate from the meaning of an invention.

101 ... push-pull converter (power converter),
102 ··· Full-bridge converter (power converter),
11 ... input end, 12 ... output end,
201, 202 ... transformer,
51, 52 ... diode (rectifying element),
7 ... Reactor,
80: Control device (power converter control device),
81: Pulse width setting unit,
82: Gate pulse signal generator,
SW1, SW2, SW3, SW4 ... switching elements.

Claims (1)

Transformers (201, 202) for converting power between an input end (11) connected to a power supply and an output end (12) connected to a load;
A plurality of switching elements (SW1, SW2, SW3, SW4) that perform switching operation so as to alternately apply positive and negative voltage pulses to the transformer;
Rectifying elements (51, 52) connected to the output path of the transformer;
A reactor (7) connected between the rectifying element and the output end;
A power converter control device applied to a power converter (101, 102) comprising:
An operation switch which is the switching element to be turned on next, and a pulse width setting section (81) for setting a pulse width which is an on time of the operation switch;
A gate pulse signal generation unit (82) that generates a gate pulse signal for operating the plurality of switching elements based on the operation switch and the pulse width set by the pulse width setting unit;
Have
The pulse width setting section, Oite before straight idle period operation of the switching element over a half cycle or period of the voltage pulse pauses, operates the switch so as to balance the positive and negative voltage applied to the transformer And set the pulse width ,
When a stop signal instructing stopping of the current switching operation and transition to the pause period is received at a timing after an intermediate time in the on period of either the positive or negative voltage pulse, simultaneously with the reception of the stop signal. Turn off the current voltage pulse,
The pulse width setting unit is a voltage pulse of the next opposite polarity so as to shift to the idle period in a state where the magnetic flux generated in the transformer is brought close to zero by application of a voltage pulse, and the last before the idle period A pulse width of a last pulse which is a voltage pulse of (1) is set to a pulse width corresponding to a time when a current voltage pulse is turned on past a middle time .

Images