JP2007020243A - Inverter power supply - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform control for eliminating the magnetization of a transformer quickly even when it takes place in the transformer of an inverter power supply. <P>SOLUTION: The inverter power supply comprises an inverter circuit, a transformer Tr, rectifier circuits DS1-2, an output modulation control circuit PWM, a circuit HD for judging magnetization by detecting an undue excitation current of the transformer Tr and outputting a magnetization distinction signal Hd, a circuit KC for forbidding output modulation control from a moment in time when the magnetization distinction signal Hd is input until half period of high frequency AC current terminates and turning the switching elements TR1-4 of the inverter circuit off, and a magnetization eliminating circuit HK for correcting output modulation control such that the output of half period on the magnetization side is lowered in order to eliminate the magnetization during a predetermined period after ending the half period when the magnetization distinction signal Hd is input. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an inverter power supply device equipped with a control for eliminating the bias magnetism of a transformer.

  In general, an inverter power supply apparatus includes an inverter circuit that converts a DC power source obtained by rectifying and smoothing a commercial power source into a high frequency alternating current using a plurality of switching elements, and a transformer that transforms the high frequency alternating current to a voltage value suitable for a load. A rectifier circuit that rectifies and supplies the transformed high-frequency alternating current to a load; an output modulation control circuit that performs output modulation control such as pulse width modulation control and frequency modulation control on the inverter circuit; And a demagnetization protection circuit for protecting the switching element. The inverter power supply device to which the present invention is directed is a welding power supply, a switching power supply, a high frequency power supply for semiconductor manufacturing, an induction heating power supply, a solar ionization or a fuel cell power supply, and the like. As described above, the pulse width modulation control (PWM) and the frequency modulation control are used for the output modulation control of the switching elements forming the inverter circuit by feeding back the load voltage value, current value, etc. to the desired values. (FM). In the following description, a case of pulse width modulation control which is typical output modulation control will be exemplified.

  Transformer bias is a major problem for inverter power supplies. Since the transformer loses its function and goes into a short-circuit load state from the bias to saturation, an overcurrent flows through the switching element of the inverter circuit, and the switching element is damaged. Biasing occurs due to various factors such as sudden changes in load, variations in characteristics of switching elements, etc., and instability of the feedback control system. In particular, recent inverter power supply devices often have an operating frequency (carrier frequency) exceeding 100 kHz, and therefore, a ferrite core with low loss at high frequency is often used for the iron core of the transformer. However, since the ferrite core has a low saturation magnetic flux density value, it immediately reaches saturation with a slight bias. For this reason, in the high-frequency inverter power supply device, it is even more important to take measures against the bias. In the following, the conventional technology for countermeasures against magnetic bias will be described.

  FIG. 4 is a block diagram of an inverter power supply apparatus equipped with a conventional technology for preventing magnetic demagnetization. The three-phase bridge rectifier DP rectifies the three-phase commercial power supply AC. The smoothing capacitor C smoothes the rectified voltage and outputs a DC voltage. A direct current power source is formed by both DP and C. The inverter circuit is formed from transistors TR1 to TR4 and freewheeling diodes D1 to D4. Here, the inverter circuit system is a full-bridge system, but a push-pull system may be used. The transistors TR1 and TR4 are on / off controlled simultaneously, and the transistors TR2 and TR3 are on / off controlled simultaneously. Therefore, in the following description, TR1 and TR4 are referred to as A set of transistors, and TR2 and TR3 are referred to as B set of transistors. The inverter circuit converts a direct current voltage into a high frequency alternating current.

  The transformer Tr transforms the voltage value of the high-frequency alternating current to a voltage value suitable for the load. When the A-group transistors TR1 and TR4 are in the on state, a positive voltage is applied to the primary winding of the transformer Tr, and the primary current Ip has a positive value. Conversely, when the B sets of transistors TR2 and TR3 are on, a negative voltage is applied to the primary winding of the transformer Tr, and the primary current Ip has a negative value. When the application time (voltage integral value) of these positive and negative voltages becomes unbalanced, the magnetization is generated. That is, if there is a large difference between the on-time of the A set of transistors TR1 and TR4 and the on-time of the B set of transistors TR2 and TR3 due to a sudden load change or the like, a magnetic bias occurs. The secondary rectifiers DS1 and DS2 rectify the transformed high-frequency alternating current. The reactor L smoothes the rectified direct current and supplies it to the load. In some cases, a capacitor may be used instead of the reactor L for smoothing.

  The voltage detection circuit VD detects the output voltage Vo and outputs a voltage detection signal Vd. The voltage setting circuit VR outputs a voltage setting signal Vr having a desired value. The error amplification circuit EA amplifies the error between the voltage setting signal Vr and the voltage detection signal Vd and outputs an error amplification signal ΔV. The sawtooth wave oscillation circuit NH outputs a sawtooth wave signal Nh having a frequency twice as high as the high frequency AC frequency of the inverter circuit. By setting the frequency of the sawtooth signal Nh, the frequency of the high frequency alternating current (carrier frequency) is set. The pulse width modulation control circuit PWM receives the error amplification signal ΔV and the sawtooth wave signal Nh, performs pulse width modulation, and outputs pulse width modulation signals Pwa and Pwb. The pulse width modulation signal Pwa is a signal for turning on the A set of transistors TR1 and TR4, and the pulse width modulation signal Pwb is a signal for turning on the B set of transistors TR2 and TR3. As will be described later with reference to FIG. 6, the pulse width modulation signals Pwa and Pwb are signals shifted by a half cycle, and both signals form one cycle of high-frequency alternating current. The current detection circuit ID detects the primary current Ip of the transformer Tr and outputs a current detection signal Id. The bias detection circuit HD receives the current detection signal Id as input, and detects the bias magnetism by detecting that the excitation current is excessive, and outputs the bias detection signal Hd, as will be described later with reference to FIG. (High level). When the demagnetization discrimination signal Hd is output, the prohibition circuit KC immediately outputs the pulse width prohibition signals Pka and Pkb that prohibit the pulse widths of the pulse width modulation signals Pwa and Pwb during the half cycle. Therefore, when the demagnetization discrimination signal Hd is not output, Pwa = Pka and Pwb = Pkb. The drive circuit DV receives the pulse width inhibition signals Pka and Pkb, and outputs a drive signal Sa for the A set of transistors TR1 and TR4 and a drive signal Sb for the B set of transistors TR2 and TR3.

  FIG. 5 is a waveform diagram of the primary current Ip and the excitation current Ir of the transformer Tr showing the method of discriminating the magnetic bias. The figure shows a half-cycle current waveform. At time t1, the A-group or B-group transistors are turned on, and the current supply starts. In the figure, a case where a magnetic bias occurs from time t2 due to a sudden change in load or the like is illustrated. As shown in FIG. 5B, when the magnetization is generated, the exciting current Ir is rapidly increased and becomes excessive. Therefore, it is possible to discriminate the magnetic bias by detecting that the value of the excitation current Ir is equal to or greater than the reference value Itr. The exciting current Ir can be calculated by Ir = Ip−K · Is. Here, K is the turn ratio of the transformer Tr, and Is is the secondary current value.

  Further, as shown in FIG. 6A, the primary current Ip has different current values in the flat portion such as L1 and L2 depending on the size of the load. However, the primary current value Ip also rises sharply with the sudden rise in the excitation current Ir and becomes equal to or higher than the reference value It at time t3. This can be detected to determine the bias. At this time, the current values of the flat portions are different between the waveforms L1 and L2, but there is no significant difference in the sudden increase in current due to the bias. For this reason, the deviation detection accuracy is not so varied. Therefore, it can be determined that the excitation current Ir is excessive by determining that the excitation current Ir or the primary current Ip is equal to or greater than a reference value.

  FIG. 6 is a timing chart of each signal in the inverter power supply apparatus described above with reference to FIG. FIG. 6A shows the sawtooth wave signal Nh, FIG. 5B shows the A set of pulse width modulation signals Pwa, FIG. 6C shows the B set of pulse width modulation signals Pwb, and FIG. The current detection signal Id, (E) in the figure shows the time variation of the demagnetization discrimination signal Hd, and (F) in the figure shows the time variation of the B sets of pulse width inhibition signals Pkb. One cycle of the sawtooth signal Nh shown in FIG. 1A is a half cycle of high-frequency alternating current. The figure shows the case where the negative voltage application time to the transformer Tr is long and unbalanced, and the current detection signal Id (= primary current Ip) shown in FIG. is there. Therefore, this is a case where the magnets are biased toward the B sets of transistors TR2 and TR3. Hereinafter, a description will be given with reference to FIG.

  During the period from time t1 to time t2, the pulse width modulation signal Pwa for the A sets of transistors TR1 and TR4 is output as shown in FIG. 5B, and current detection is performed as shown in FIG. The signal Id has a positive value. During this period, a positive voltage is applied to the transformer Tr. During the period from time t2 to t3, as shown in FIG. 5C, the pulse width modulation signal Pwb for the B sets of transistors TR2 and TR3 is output, and as shown in FIG. The signal Id has a negative value. During this period, a negative voltage is applied to the transformer Tr.

  When a bias magnetism occurs due to a sudden change in load during the period from time t4 to time t5, as shown in FIG. 5D, the value of the current detection signal Id increases rapidly as described above with reference to FIG. When the value of the current detection signal Id becomes equal to or greater than the reference value It at time t41, as shown in FIG. 5E, the magnetic bias discrimination signal Hd changes to the high level until time t5. In response to this, the group B pulse width modulation signal Pwb shown in FIG. 11C is prohibited to the low level at time t41, and the group B pulse width inhibition signal Pkb shown in FIG. . As a result, the current value is further increased to prevent the B sets of transistors TR2 and TR3 from being damaged. However, in the countermeasure against the demagnetization by the pulse width inhibition control, the demagnetization state is continued in order to inhibit the pulse width in a state where the demagnetization has progressed considerably. As a result, the current value suddenly rises and the pulse width prohibition tends to occur repeatedly in the half periods on the B side from time t6 to t7, time t8 to t9, and time t10 to t11. In such a state, an overcurrent is continuously applied to the transistor to energize it, and although it is not damaged, the reliability is lowered. Furthermore, since the continuous state of overcurrent also becomes an output unstable state, the system reliability of the inverter power supply device is also lowered.

  In addition to the above-mentioned bias magnetism countermeasure technology, the following conventional technologies are disclosed. The first is a technique for detecting a direct current component of the primary current Ip of the transformer with a filter and discriminating the magnetic bias. The second is a technique for integrating the primary voltage applied to the transformer and discriminating the magnetic bias based on the value. When the demagnetization is determined, the pulse width modulation signal is corrected so as to eliminate the demagnetization (see, for example, Patent Documents 1 to 3).

Japanese Patent Publication No. 3-43938 Japanese Patent No. 2973564 JP-A-8-187575

  In the prior art in which the pulse width is prohibited when the above-described primary current value is equal to or greater than the reference value, the biased state cannot easily be eliminated, so that the pulse width is easily prohibited. For this reason, the reliability of the switching element of the inverter circuit is lowered, and the system reliability is lowered due to an unstable output state. Next, in the conventional technique for detecting the direct current component of the primary current, the switching element cannot be protected from abrupt occurrence of demagnetization due to the time delay of the direct current component detection filter. As described above, since a ferrite core is often used in recent inverter power supply devices, the saturation magnetic flux density is small. fall into disrepair. Therefore, the time delay must be as short as possible. Next, in the conventional technique for integrating the primary voltage, a slight time delay occurs. Furthermore, since the primary voltage is as high as about 300 V, the voltage detection circuit becomes large and expensive. In order to detect the primary voltage, an auxiliary winding is wound around the transformer to detect the voltage. Even in this case, the voltage value is about 100 V, which is difficult to detect.

  Therefore, the present invention provides an inverter power supply device that can detect bias with simple means without time delay to prevent overcurrent and simultaneously perform control to eliminate bias.

In order to solve the above-described problem, the first invention includes an inverter circuit that converts a DC power source into high-frequency alternating current using a plurality of switching elements, a transformer that transforms the high-frequency alternating current into a voltage value suitable for a load, By rectifying this transformed high-frequency alternating current and supplying it to the load, an output modulation control circuit for controlling the output modulation of the inverter circuit, and detecting that the excitation current of the transformer is excessive A demagnetization discriminating circuit that discriminates the magnetism of the transformer and outputs a demagnetization discrimination signal, and prohibits the output modulation control until the half cycle of the high-frequency alternating current ends after the input of the demagnetization discrimination signal In an inverter power supply device comprising a prohibition circuit that changes the switching element of the inverter circuit to an off state,
The output modulation control is performed so that the half-cycle output on the side that is demagnetized decreases for a predetermined period from the end of the half-cycle when the demagnetization determination signal is input to eliminate the demagnetization. An inverter power supply device is provided with a demagnetization elimination circuit to be corrected.

  According to a second aspect of the present invention, the excitation current of the transformer in the magnetic field discrimination circuit according to the first aspect of the invention is excessive, and the rate of increase excluding the rising period of the primary current of the transformer and / or An inverter power supply device that detects when the value of the primary current is equal to or greater than a reference value.

  In the third invention, the method of correcting the output modulation control is demagnetized so that the output of the demagnetized half-cycle in the demagnetization circuit described in the first invention is reduced. The inverter power supply apparatus is characterized in that the output modulation control is modified so as to shorten a pulse width of a half cycle on the side.

  According to the present invention, when the demagnetization is detected without time delay due to the excessive excitation current, the switching element of the inverter circuit is immediately turned off to prevent the damage. Subsequently, the bias can be eliminated by reducing the output on the side that is biased for a predetermined period from the bias determination. For this reason, the continuous state of the overcurrent accompanying the continuation of the biased state can be prevented, and the reliability of the transistor and the system can be improved. Furthermore, since the bias can be determined by current detection, the detection means is easy and inexpensive.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a block diagram of an inverter power supply apparatus according to an embodiment of the present invention. In the figure, the same blocks as those in FIG. 4 described above are denoted by the same reference numerals, and description thereof is omitted. Hereinafter, blocks indicated by dotted lines different from those in FIG. 4 will be described.

  This figure is obtained by adding a demagnetization elimination circuit HK to FIG. This demagnetization canceling circuit HK is a half-demagnetization side in order to cancel the demagnetization for a predetermined demagnetization canceling period Th from the time when the half cycle in which the demagnetization determination signal Hd is input ends. The pulse width modulation signal Pwa or Pwb is corrected so that the output of the period is lowered, and the pulse width correction signals Pha and Phb are output.

  FIG. 2 is a timing chart of each signal in the inverter power supply apparatus of FIG. 1 described above. FIG. 6A shows the sawtooth wave signal Nh, FIG. 5B shows the A set of pulse width modulation signals Pwa, FIG. 6C shows the B set of pulse width modulation signals Pwb, and FIG. (E) of the current detection signal Id is a demagnetization discrimination signal Hd, (F) is a B set of pulse width correction signals Phb, and (G) is a B set of pulse width inhibition signals Pkb. Shows time change. This figure corresponds to FIG. 6 described above, and is obtained by adding B sets of pulse width correction signals Phb shown in FIG. In the figure, similarly to the case of FIG. 6, one cycle of the sawtooth signal Nh shown in FIG. The figure shows the case where the negative voltage application time to the transformer Tr is long and unbalanced, and the current detection signal Id (= primary current Ip) shown in FIG. is there. Therefore, this is a case where the magnets are biased toward the B sets of transistors TR2 and TR3. Hereinafter, a description will be given with reference to FIG.

  When the transformer is demagnetized at time t41 and the value of the primary current Ip becomes equal to or higher than the reference value It, as shown in FIG. The signal Hd is output (High level). The bias is generated on the side where the primary current Ip has a negative value (the side of the B sets of transistors TR2 and TR3). In response to this, the pulse width inhibition signal Pkb on the B side is immediately prohibited to the Low level as shown in FIG. As a result, the B sets of transistors TR2 and TR3 are immediately turned off, and the primary current Ip is prohibited from flowing beyond the reference value It as shown in FIG. Thereby, the breakage of the B sets of transistors TR2 and TR3 can be prevented.

  At subsequent time t5, when the half cycle in which the demagnetization determination signal Hd shown in FIG. 4D is output (High level) ends and becomes Low level, a predetermined demagnetization elimination period Th from that time (t5). Starts. Here, Th = 2 periods. Th = 1 to about 10 cycles is an appropriate range. In the demagnetization canceling engine Th, as shown in FIG. 4F, the pulse width of the demagnetized B-side half cycle (t6 to t7 and tt8 to t9) is shortened to reduce the B-side half cycle. Try to reduce the output. That is, the B sets of pulse width modulation signals Pwb shown in (C) in the period t6 to t7 are corrected to the pulse width correction signals Phb shortened by time t62, as shown in (F). Is done. Similarly, the B sets of pulse width modulation signals Pwb shown in FIG. 8C during the period from time t8 to time t9 are changed to pulse width correction signals Phb shortened by time t82 as shown in FIG. Will be corrected. This pulse width shortening method is performed by reducing the pulse width of the original pulse width modulation signal Pwb to a pulse width obtained by multiplying the pulse width by a coefficient (0 <coefficient <1.0). There is also a method of correcting to a predetermined minimum pulse width.

  During the demagnetization elimination period Th, the demagnetization toward the B side is eliminated by lowering the output of the half-cycle on the B side that is demagnetized. In this embodiment, since the output modulation control method is a pulse width modulation control method, the pulse width may be shortened in order to reduce the output. When the output modulation control method is a frequency modulation control method (with a constant pulse width), the frequency may be lowered in order to reduce the output.

  In order to determine the magnetic bias, it is only necessary to detect that the excitation current Ir of the transformer has become excessive, as described above with reference to FIG. As this method, as shown in FIG. 5, there is a method of detecting that the value of the primary current Ip or the exciting current Ir is equal to or higher than a reference value. FIG. 3 is a current waveform diagram showing a method for detecting an excessive excitation current other than this. FIG. 4A shows a half-cycle waveform of the primary current Ip, and FIG. 4B shows a differential value Bi = dIp / dt of the primary current Ip. As shown in FIG. 5B, the differential value Bi indicates the rate of increase of the primary current Ip, and the bias is discriminated at time t3 when the value becomes equal to or higher than the reference differential value Bt. This method based on the rate of increase may be able to reliably detect the occurrence of magnetic bias earlier than the method of comparing the primary current value Ip with the reference value It. Detection can be performed using both, and logical product can be taken to make detection accuracy more reliable.

1 is a block diagram of an inverter power supply device according to an embodiment of the present invention. It is a timing chart of each signal of the inverter power supply device of FIG. It is a wave form diagram of primary current Ip of a transformer which shows a demagnetization discrimination method, and its differential value Bi. It is a block diagram of the inverter power supply device in a prior art. It is a wave form diagram of primary current Ip and exciting current Ir of a transformer which shows a demagnetization discrimination method. It is a timing chart of each signal of the inverter power supply device of FIG.

Explanation of symbols

AC commercial power supply Bi differential value Bt reference differential value C smoothing capacitor D1 to D4 freewheeling diode DP three-phase bridge rectifier DS1, DS2 secondary rectifier DV drive circuit EA error amplification circuit HD demagnetization determination circuit Hd demagnetization determination signal HK demagnetization cancellation Circuit ID Current detection circuit Id Current detection signal Ip Primary current Ir Excitation current Is Secondary current It Reference value Itr Reference value KC Forbidden circuit L Reactor NH Sawtooth wave oscillation circuit Nh Sawtooth wave signal Pha, Phb Pulse width correction signals Pka, Pkb Pulse width inhibition signal PWM Pulse width modulation control circuit Pwa, Pwb Pulse width modulation signal Sa, Sb Drive signal Th Demagnetization elimination period Tr Transformer TR1-TR4 Transistor VD Voltage detection circuit Vd Voltage detection signal Vo Output voltage VR Voltage setting circuit Vr Voltage setting signal ΔV Error amplification signal

Claims (3)

An inverter circuit that converts a DC power source into a high-frequency alternating current using a plurality of switching elements, a transformer that transforms the high-frequency alternating current into a voltage value suitable for the load, and a rectifier that rectifies the high-frequency alternating current that is supplied to the load. A circuit, an output modulation control circuit for controlling output modulation of the inverter circuit, and detecting that the magnetizing current of the transformer is excessive, thereby determining the magnetism of the transformer and outputting a magnetism discrimination signal And the output modulation control is prohibited and the switching element of the inverter circuit is changed to the OFF state until the half cycle of the high-frequency alternating current is completed after the input of the demagnetization determination signal. In an inverter power supply device comprising a circuit,
The output modulation control is performed so that the half-cycle output on the side that is demagnetized decreases for a predetermined period from the end of the half-cycle when the demagnetization determination signal is input to eliminate the demagnetization. An inverter power supply apparatus comprising a demagnetization canceling circuit for correction.   The fact that the exciting current of the transformer in the magnetic bias discrimination circuit according to claim 1 has become excessive is that the rate of increase and / or the value of the primary current excluding the rising period of the primary current of the transformer is a reference value. An inverter power supply device, characterized in that it is detected by the above. 2. The method of correcting the output modulation control so that the half-cycle output on the demagnetized side in the demagnetization circuit according to claim 1 is reduced, and the pulse width of the half-cycle on the demagnetized side is shortened. The inverter power supply apparatus is characterized in that the output modulation control is corrected as described above.

Images