JP2007124842A - Inverter - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce the power loss of an inverter having bi-directional switching elements. <P>SOLUTION: A dead time management portion 105 performs the adaptive control of the length of dead time for each phase when a pair of semiconductor switching elements are both put into an OFF state simultaneously based on the direction of currents I<SB>U</SB>, I<SB>V</SB>, I<SB>W</SB>and middle point voltages V<SB>MU</SB>, V<SB>MV</SB>, V<SB>MW</SB>. For instance, when the current I<SB>U</SB>of U-phase is supplied from the semiconductor switch SWUH to a load 10, its gate voltage is operated via a gate driver 106 so as to put the element SWUH first into an OFF state and simultaneously, the middle point voltage V<SB>MU</SB>is measured. If the middle point voltage V<SB>MU</SB>starts to drop after that even during the dead time period, it can be considered that the change to the OFF state of the element SWUH has been completed. Its gate voltage is operated at that time so that the switching element SWUL at the low voltage side is put into an ON state. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

The present invention relates to an inverter configuration, and more particularly to a means for reducing power loss.
The present invention is very useful for reducing power loss in an inverter and simplifying the configuration of the main circuit of the inverter.

(Technical background)
In conventional inverter technology for inductive loads such as electric motors, it is considered indispensable to connect a flywheel diode in parallel to the switching element, and the main circuit has a flywheel diode. There has never been a report that no inverter has been realized. This is because a motor or the like has a large inductance in an electric circuit, and therefore when a current is changed, a voltage (= the product of a time differential value of the current and the inductance) is generated between the terminals. .

  Assume that in an inverter circuit using an electric motor as a load, a dead time is provided to prevent a short-through current, assuming that there is a circuit that does not have a flywheel diode connected in parallel as a switching element. Then, one switching element of a certain arm remains in a non-conductive state, and the other switching element is switched from a conductive state to a non-conductive state. The shorter the time taken to switch between the two states, the greater the time differential value of the motor current and the greater the voltage generated between the motor terminals. The voltage generated between the motor terminals is applied to the midpoint of the inverter arm. If the voltage generated between the motor terminals becomes too large, the breakdown voltage of the switching element constituting the main circuit of the inverter will be exceeded, and the switching element may break down and cause permanent destruction.

On the other hand, consider a case where an inverter circuit having a motor as a load has a flywheel diode in parallel with the switching element. At this time, if a dead time is provided to prevent a short-through current, the other switching element is switched from the conducting state to the non-conducting state while one switching element of a certain arm is in the non-conducting state. become.
Therefore, in this case, the forward voltage is applied to the flywheel diode due to the voltage generated between the terminals of the motor, and a regenerative current flows, thereby reducing the time differential value of the motor current, and as a result, The voltage generated between the terminals is reduced. Thus, in an inverter circuit using a motor as a load, if a flywheel diode is provided in parallel with the switching element, the voltage generated between the motor terminals is reduced, so that the switching element does not exceed the withstand voltage. It is possible to eliminate the fear of permanent destruction without breakdown. This is the reason why it is considered essential to connect a flywheel diode in parallel to the switching element in the inverter circuit.

(Prior art closest to this application)
By the way, in a bidirectional switching element formed of, for example, a unipolar type semiconductor or the like, a current can flow in both positive and negative directions in the on state. Such a semiconductor switching element does not have a certain relatively large on voltage as in the case of a junction type switching element, and the voltage drop during conduction (on voltage) is a relatively small on resistance and current. Determined by the product of

On the other hand, even in the case of flywheel diodes even if the elements are made of the same semiconductor material, these are generally Schottky junction type, PN junction type, or PIN junction type diodes, and therefore flow through them. The relationship between current and on-voltage is approximately exponential. This exponential characteristic can be generally regarded as having a constant on voltage regardless of the magnitude of the current as a first order approximation.
On the other hand, in the case of the bidirectional semiconductor switching element as described above, since the element is basically of a resistance modulation type, the switching element has a constant value as in the above junction type element. Therefore, the on-voltage drop (on voltage) is determined by the product of the on-resistance and the current, as described above.

For this reason, a flywheel having a high regenerative current by increasing the proportion of current flowing from the load toward the DC power source (hereinafter referred to as “regenerative current”) to the bidirectional switching element. The power loss can be greatly reduced as compared to the case where only the diode flows.
As described above, an inverter in which a switching element is configured by a bidirectional switching element has a characteristic that power loss due to a regenerative current is basically small.
Transistor technology, March 2004 issue

  However, during the dead time period provided to prevent the occurrence of the shoot-through current due to the power supply short-circuit state, the switching elements on both the high potential side and the low potential side are simultaneously in the off state. The regenerative current cannot be shunted to the switching element side. That is, during the dead time period, the regenerative current flows exclusively through the flywheel diode having a high on voltage, and at this time, the problem of on loss due to the regenerative current also appears.

In addition, when the above regenerative current is applied to the flywheel diode, the reverse current (that is, the known reverse current) is applied for a short time after the voltage applied to the flywheel diode is reversed from the forward direction to the reverse direction. Direction recovery current) flows. Since the reverse recovery current flowing through the flywheel diode temporarily induces a power supply short-circuit state through the other semiconductor switching element in the pair, this causes a large switching loss.
At this time, the reverse recovery current increases as the regenerative current flowing through the flywheel diode increases. Therefore, even if the dead time is short, if the regenerative current flowing through the flywheel diode is large, the reverse recovery current also increases accordingly, which leads to an increase in power loss that cannot be ignored. In addition, the power supply short-circuit state may place a heavy burden on the semiconductor switching element on the short-circuit path. Therefore, it is necessary to design for these viewpoints in consideration of reliability and durability.

  Recently, high-temperature operability for inverters has been strongly demanded. For example, when a semiconductor switching element using a GaN crystal or a diode using a GaN crystal is used for the main circuit of the inverter, A high-temperature operating environment of about 250 ° C., which greatly exceeds the current 120 ° C., is entering an era in which the design conditions of the inverter can be assumed. However, according to such a high-temperature-oriented circuit configuration, the on-voltage of the flywheel diode becomes significantly higher than the conventional circuit configuration in which a junction diode made of, for example, Si is used for the flywheel diode. . For this reason, the above-mentioned on-loss problem associated with regenerative current may develop into a more serious problem in the future.

The present invention has been made to solve the above-described problems, and an object thereof is to further reduce power loss in an inverter having a bidirectional semiconductor switching element.
Another object of the present invention is to reduce the size and cost of the inverter.

In order to solve the above problems, the following means are effective.
That is, the first means of the present invention has a bidirectional semiconductor switching element capable of flowing a current bidirectionally in a controlled current path to be controlled for opening and closing, and the controlled current path In the inverter provided with a pair of semiconductor switching elements connected in series for each phase, a midpoint voltage measuring means for measuring the voltage at the connection point of the pair of semiconductor switching elements for each phase; The current detection means for detecting the current direction of each phase and the length of the dead time during which both the pair of semiconductor switching elements are simultaneously turned OFF are set to the above current direction and the above voltage for each phase. A dead time management means for adaptively controlling based on the switching time, and the dead time management means for switching the pair of one semiconductor switching element from the ON state to the OFF state. When the increase or decrease in the voltage from the time when the FF command signal is output indicates a displacement amount exceeding a predetermined threshold (> 0), the other semiconductor switching element of the pair is turned from the OFF state to the ON state. Is to output an ON command signal for switching to.

However, any number of the above phases may be provided. The present invention can be applied to a single-phase or multi-phase inverter having any number of phases. Further, the current detection means may be any means that can detect the direction of the current of each phase, and does not necessarily need to be a means for accurately obtaining the current value.
Further, the DC voltage of the DC power source to be used may be arbitrary. For example, an inverter with a relatively high power supply voltage of about several hundred volts or more is used according to the present invention by using a semiconductor switching element having a high breakdown voltage and a low on-resistance constructed using a wide band gap semiconductor such as GaN. It is also possible to produce

Also, the second means of the present invention is the dead time management means of the first means, when the displacement amount reaches a value not less than 10% and not more than 20% of the DC power supply voltage used. A first voltage increase / decrease determination means for determining that the above-described ON command signal should be output is provided.
However, this threshold value can be optimized according to the characteristics of the semiconductor switching element to be used, the characteristics of the load, the power supply voltage, the number of phases of the inverter, or the like.

According to a third means of the present invention, in the dead time management means of the first or second means, the displacement amount is set to a value more than twice the voltage drop amount when the semiconductor switching element is conductive. A second voltage increase / decrease determining means for determining that the above-described ON command signal should be output at the time of arrival is provided.
However, more preferably, the ON command signal may be output when the amount of displacement reaches a value of about 4 to 8 times the voltage drop when the semiconductor switching element is conductive. This threshold value can be optimized according to the characteristics of the semiconductor switching element to be used, the characteristics of the load, the power supply voltage, the number of phases of the inverter, or the like.

According to a fourth means of the present invention, in the dead time management means of any one of the first to third means, a period measuring means for defining an upper limit value of the dead time length is provided. It is.
Such period measuring means can be configured using, for example, a computer timer or a computer dynamic step counter.
By the above means of the present invention, the above-mentioned problem can be effectively or rationally solved.

The effects obtained by the above-described means of the present invention are as follows.
That is, according to the first means of the present invention, even within the dead time period, based on the adaptive control of the dead time management means, the regenerative regenerative power that was flowing exclusively through the flywheel diode in the conventional configuration. All of the current can be passed through the semiconductor switching element. This also makes it possible to eliminate the flywheel diode from the circuit configuration of the main circuit of the inverter.

  In this case, instead of eliminating the amount of regenerative current flowing through the flywheel diode, the entire regenerative current flows through the semiconductor switching element, so that on loss due to the regenerative current of the semiconductor switching element newly occurs. However, the on-loss generated in the semiconductor switching element is greatly reduced compared to the on-loss generated in the conventional flywheel diode, and as a result, the inverter loss can be greatly reduced. In addition, by setting an appropriate dead time, it is possible to effectively prevent the occurrence of a shoot-through current due to a power supply short-circuit state at the same time.

Further, according to the first means of the present invention, since the flywheel diode itself can be eliminated, the reverse recovery current itself that flows when the voltage applied to the flywheel diode is reversed from the forward direction to the reverse direction is also entirely displayed. Can be eliminated. For this reason, according to the 1st means of this invention, the switching loss resulting from the reverse direction recovery current can also be excluded reliably.
Further, according to the first means of the present invention, since the flywheel diode itself can be eliminated, it is possible to reliably realize downsizing and cost reduction of the apparatus.

When managing the dead time, when the OFF state of both the pair of semiconductor switching elements is just realized, an ON command signal is sent to the semiconductor switching element that has been in the OFF state so far. However, according to the second or third means of the present invention, each of the corresponding threshold values is optimized so that the ON command signal is output at such a desirable timing. Can be set.
Therefore, according to the second or third means of the present invention, it is possible to optimally and surely realize an inverter with less power loss and higher efficiency than conventional. Moreover, since the burden on the semiconductor switching element is reduced by optimizing the above threshold value, the reliability and durability of the inverter can be ensured.

  Further, according to the fourth means of the present invention, it is possible to reliably complete the dead time even when there is a measurement error or erroneous detection due to noise or the like. For this reason, a more reliable inverter can be comprised.

Hereinafter, the present invention will be described based on specific examples.
However, the embodiments of the present invention are not limited to the following examples.

FIG. 1 shows a circuit diagram of the inverter 100 according to the first embodiment, and FIG. 2 shows a control block diagram of the inverter 100. The six semiconductor switching elements SWUH, SWVH, SWWH, SWUL, SWVL, SWWL constituting the main circuit 101 of the inverter 100 are all GaN-HEMTs (High Electron Mobility Transistors), that is, GaN-based FETs (field effect transistors). A pair of semiconductor switching elements, each having a controlled current path connected in series, are provided for each phase (U phase, V phase, W phase) of the three-phase motor 10. The connection point of each pair of semiconductor switching elements in each phase is hereinafter referred to as a midpoint, and the voltage at each midpoint of each phase (U phase, V phase, W phase) is hereinafter referred to as a midpoint voltage. _They are called V _MU , midpoint voltage V _MV , and midpoint voltage V _MW . The output of the inverter 100, that is, the currents I _U , I _V , and I _W of each phase flowing through the three-phase motor 10 can be controlled by switching the midpoint voltage between these levels at appropriate times.

  A low-pass filter may be provided between the inverter 100 and the load (three-phase motor 10). Here, the frequency of the rectangular wave in the chopper control of the PWM circuit of the inverter 100 is assumed to be about several kHz to several tens of kHz, and the frequency of the alternating current to be supplied to the load is assumed to be about 10 to 240 Hz. is doing.

Each phase midpoint voltage measurement unit 102 measures the above midpoint voltages V _MU , V _MV , and V _MW as needed. Each phase current measurement unit 103 measures currents I _U , I _V , and I _W of each phase as needed. Each phase midpoint voltage high / low command signal generation unit 104 is constituted by a PWM circuit, and outputs the above midpoint voltage command value for each phase. Hereinafter, this command value is referred to as a high / low command signal S.

The dead time management unit 105 corresponds to the dead time management means according to claim 1 of the present invention, and determines the length of the dead time during which both the pair of semiconductor switching elements are simultaneously turned off for each phase. In addition, adaptive control is performed based on the directions of the currents I _U , I _V and I _W and the midpoint voltages V _MU , V _MV and V _MW . This part may be constituted by hardware (electric circuit) or software that operates on a computer.
Further, the gate driver 106 controls each gate of each of the semiconductor switching elements to be in an ON state or an OFF state in accordance with a command (predetermined command signal) issued by the dead time management unit 105.

  In the following, the logic of gate control of the semiconductor switching element performed in order to flow all the regenerative current within the dead time period to the semiconductor switching element in the inverter 100 will be described with reference to FIG. This gate control is divided into four cases (cases 1 to 4 in Table 1) according to the operation direction of each midpoint voltage given by the above-described high and low command signal S and the direction of the load current of each phase (positive and negative). Each can be executed separately.

In Table 1, the current in the direction from the inverter 100 toward the load (three-phase motor 10) is a positive current. Since the following description regarding the logic of gate control is common to each phase, the following description will be made assuming that the first phase in FIG. 3 is the U phase.

(Case 1)
Midpoint voltage operation direction: High ⇒ Low
Load current: positive The operation (gate control) in this case is an operation to switch the semiconductor switching element SWUH on the high potential side to the ON state so that the semiconductor switching element SWUL on the low potential side is turned on. .

Here, first, the gate voltage is operated via the gate driver 106 (: high potential SW off operation a1) so that the high potential side semiconductor switching element SWUH is turned off, and at the same time, the midpoint voltage _VMU is Measure (: midpoint voltage drop determination b1). If reduction of the so-called midpoint even within the dead time period the voltage V _MU starts, since the change of the semiconductor switching devices SWUH the OFF state of the high potential side it can be regarded as complete, at which time The gate voltage is manipulated through the gate driver 106 so that the low potential side semiconductor switching element SWUL is turned on (low potential side SW on operation e1).

As a guideline for lowering the midpoint voltage _VMU at this time, it is appropriate to set the power supply voltage to about 10 to 20%. By doing this, it is possible to reduce the time (dead time period) during which both the high potential side and low potential side semiconductor switching elements (SWUH, SWUL) are in the OFF state simultaneously, and the low potential side semiconductor switching element. when the element SWUL is conducting, the load current I _U is supplied to the load through the low-potential side semiconductor switching element SWUL (3-phase motor 10).
And by this processing method, generation | occurrence | production of the shoot through current resulting from a power supply short circuit state can be prevented. Further, since the conventional flywheel diode is not provided, the occurrence of the power supply short-circuit state due to the above-described conventional reverse recovery current is prevented, and the loss is reduced accordingly.

  Note that dead time period measurements c1, c2, c3, and c4 in FIG. 3 constitute the period measurement means of the present invention, and even when there is a measurement error or erroneous detection due to noise or the like, this period measurement is performed. The dead time can be surely completed by the action. OR determinations d1, d2, d3, and d4 in FIG. 3 are timing determinations from either the midpoint voltage determination processing (b1, b2, b3, b4) or the dead time period measurement (c1, c2, c3, c4). It consists of logical operation processing or logical sum circuit that adopts the result. However, when one is adopted, the other is ignored until the next start of the program or does not operate until the next start of the program.

(Case 2)
Midpoint voltage operation direction: High ⇒ Low
Load current: negative The operation (gate control) in this case is also an operation for switching the semiconductor switching element SWUH on the high potential side to the ON state so that the semiconductor switching element SWUL on the low potential side is turned on. .

Here, first semiconductor switching devices SWUH high potential side operates the gate voltage so that the OFF state (: high potential side SW off operation a2), to measure the midpoint voltage V _MU simultaneously (: midpoint voltage rise Decision b2). In this case, since the load current is negative, a regenerative current flows through the semiconductor switching element SWUH on the high potential side. Since the semiconductor switching devices SWUH the high potential side from the state regenerative manner was flowing in the OFF state, the midpoint voltage V _MU from the time it is in the OFF state starts to rise. Therefore, it can be regarded as the change to the OFF state of the high potential side of the semiconductor switching devices SWUH is completed if elevated even midpoint voltage V _MU be within the period of the so-called dead time Hajimare. When the rise is detected, the gate voltage is manipulated so that the low-potential side semiconductor switching element SWUL is turned on (low-potential side SW-on operation e2).

As a measure of the rise of the midpoint voltage _VMU at this time, it is appropriate to set it to several times the voltage drop (on voltage) when the semiconductor switching element is conductive. By doing so, the semiconductor switching element SWUL on the low potential side can be conducted in a short time after the semiconductor switching element SWUH on the high potential side is turned off. As a result, a regenerative current (negative load) Current) can continue to flow smoothly through the semiconductor switching element SWUL on the low potential side.

  And by this processing method, generation | occurrence | production of the big on loss by the high on voltage of the flywheel diode with which the conventional method was equipped can be suppressed, preventing generation | occurrence | production of the shoot through current resulting from a power supply short circuit state. Further, since the flywheel diode itself is not installed, the above-described conventional reverse recovery current is eliminated, the occurrence of a power supply short-circuit state resulting therefrom is prevented, and the switching loss corresponding to that is suppressed.

(Case 3)
Midpoint voltage operation direction: Low ⇒ High
Load current: positive In this case, the operation (gate control) is an operation for switching the semiconductor switching element SWUL on the low potential side to the ON state so that the semiconductor switching element SWUH on the high potential side is turned on. .

Here, first low-potential-side semiconductor switching element SWUL operates the gate voltage so that the OFF state (: low potential side SW off operation a3), to measure the midpoint voltage V _MU simultaneously (: midpoint voltage falling Determination b3). In this case, since the load current I _U is positive, a regenerative current (I _U ) flows through the semiconductor switching element SWUL on the low potential side. Since the state where the regenerative manner was flowing in the semiconductor switching element SWUL OFF state of the low potential side, the midpoint voltage V _MU is lowered. Therefore, can be regarded as the change to the OFF state of the low potential side of the semiconductor switching element SWUL is completed if a so-called dead time of the midpoint voltage V _MU even within a period of lowered Hajimare. Therefore, when the falling began the midpoint voltage V _MU, as semiconductor switching devices SWUH the high potential side is turned ON to operate the gate voltage via a gate driver 106 (: high potential side SW ON Operation e3).

As a guideline for lowering the midpoint voltage _VMU , it is appropriate to set it to be several times the voltage drop (on voltage) when the semiconductor switching element is conductive. Thereby, the low-potential side semiconductor switching element SWUL conducts the high potential side of the semiconductor switching devices SWUH within a short time from when the OFF state, the load current I _U, through the semiconductor switching devices SWUH the high potential side Smoothly supplied to the load.

  And by this processing method, generation | occurrence | production of the big on loss by the high on voltage of the flywheel diode with which the conventional method was equipped can be suppressed, preventing generation | occurrence | production of the shoot through current resulting from a power supply short circuit state. Further, since the flywheel diode itself is not installed, the above-described conventional reverse recovery current is eliminated, the occurrence of a power supply short-circuit state resulting therefrom is prevented, and the switching loss corresponding to that is suppressed.

(Case 4)
Midpoint voltage operation direction: Low ⇒ High
Load current: negative The operation (gate control) in this case is also an operation for switching the semiconductor switching element SWUL on the low potential side so that the semiconductor switching element SWUH on the high potential side is turned on. .

Here, first low-potential-side semiconductor switching element SWUL operates the gate voltage so that the OFF state (: low potential side SW off operation a4), to measure a midpoint voltage V _MU simultaneously (: midpoint voltage rise Determination b4). Since it can be regarded as a change to the low potential side of the semiconductor switching element SWUL OFF state is completed if rises the midpoint voltage V _MU be within the period of the so-called dead time starts, at that time, a high potential The gate voltage is manipulated via the gate driver 106 so that the semiconductor switching element SWUH on the side is turned on (high potential side SW on operation e4).

As a guideline for the rise of the midpoint voltage _VMU , it is appropriate to set it to about 10 to 20% of the power supply voltage. Thereby, the low-potential side semiconductor switching element SWUL is rendered conductive semiconductor switching device SWUH of the high-potential-side in a short time from when the OFF state, the regenerative amperometric I _U semiconductor switching element having a high potential side SWUH It can continue to flow smoothly through.

  And by this processing method, generation | occurrence | production of the big on loss by the high on voltage of the flywheel diode with which the conventional method was equipped can be suppressed, preventing generation | occurrence | production of the shoot through current resulting from a power supply short circuit state. Further, since the flywheel diode itself is not installed, the above-described conventional reverse recovery current is eliminated, the occurrence of a power supply short-circuit state resulting therefrom is prevented, and the switching loss corresponding to that is suppressed.

FIG. 4 illustrates a control procedure for concretely executing the above gate control logic (gate operation). Since the above description regarding the logic of gate control is common to each phase, the following description will be made assuming that the first phase in FIG. 3 is the U phase.
The program 200 in FIG. 4 is for executing processing for the first phase of the dead time management unit 105 in FIG. Here, the above-described high / low command signal S is defined by the following 2-bit signal.
(High / Low command signal S)
S = 00: High ⇒ Low (The SWUH is in the ON state.
Operation command to switch SWUL to ON state)
S = 10: Low ⇒ High (If SWUL was in the ON state,
Operation command to switch SWUH to ON state)

The program 200 is activated from the above-described midpoint voltage high / low command signal generation unit 104 at the time when these switching operations become necessary. At that time, first, in the first step 210, the current current I _U in the U phase is input from each phase current measuring unit 103. Next, in step 220, the direction of the current I _U is determined. If I _U <0, the process proceeds to step 225, and if not, the process proceeds to step 230.

In step 225, a logical sum of a 2-bit mask pattern mp for logical sum operation defined as follows and the above high / low command signal S is calculated corresponding to each bit, and the result is stored in a variable K. To do. This variable K is a 2-bit variable used as one of the work areas of the program 200.
(Mask pattern)
mp: 01
As a result, as shown in Table 1, the variable K stores the following contents for each of the cases 1 to 4 described above. And the case division part 1050 of FIG. 3 is realizable by the above process (steps 210-225).
(1) Case 1: K = 0
(2) Case 2: K = 1
(3) Case 3: K = 2
(4) Case 4: K = 3

In step 230, enter the midpoint voltage V _MU each phase midpoint voltage measuring unit 102, the next step 235, saves the value to the variable V _0. In step 240, it reads the current time t from the timer saves that value in a variable T _0. These saving processes (steps 230 to 240) are for storing the initial state at the start of the dead time period. By these saving processes, “initial state saving” i1, i2 in FIG. , I3, i4 can be realized simultaneously.

  Next, in step 245, command AU (K) is executed according to the contents shown in Table 2 below. This command AU (K) (0 ≦ K ≦ 3) is stored in the form of an array in a predetermined storage area, and each of the commands AU (K) (0 ≦ K ≦ 3) indicating each predetermined process corresponding to the variable K described above. Each command code (stored code) is held. Thereby, for example, when K = 0, the command code AU (0) is selected, and the operation of turning off the semiconductor switching element SWUH is executed via the gate driver 106.

なお、この処理は、図３の「オフ操作」ａ１，ａ２，ａ３，ａ４の各制御処理に相当している。

This process corresponds to each control process of “OFF operation” a1, a2, a3, a4 in FIG.

Next, in step 250, the midpoint voltage V _MU is input from each phase midpoint voltage measurement unit 102. In step 255, determination process 1 shown in the following equation (1) is performed. If this relationship is established, the process proceeds to step 265, and if not, the process proceeds to step 260.
(Judgment process 1)
| V _MU −V ₀ |> D (K) (1)
Here, the threshold value D (K) is defined as follows in the form of an array in a predetermined storage area.
(Definition of array D (K))
D (0) = α (α is a value of 15% of the power supply voltage)
D (1) = β (β is a value that is four times the voltage drop when SWUH is energized)
D (2) = β
D (3) = α
This process (steps 250 to 255) corresponds to determination processes b1, b2, b3, and b4 related to the midpoint voltage in FIG. In other words, the first voltage increase / decrease determination unit and the second voltage increase / decrease determination unit of the present invention are simultaneously realized by step 255 according to such an array definition.

Next, in step 260, the determination process 2 shown in the following equation (2) is performed. If this relationship is established, the process proceeds to step 265, and if not, the process proceeds to step 250.
(Judgment process 2)
t−T ₀ > ΔT (2)
Here, ΔT is a threshold value that defines an upper limit value of the length of the dead time, and may be about 3 μm to 5 μsec, for example. Further, for example, each case may be individually defined in the same manner as the threshold value D (K) described above.
The dead time period measurement c1, c2, c3, c4 in FIG. 3 corresponding to the period measurement means of the present invention is composed of the above-mentioned steps 240 and 260, and defines the upper limit value of the dead time length. The OR determinations d1, d2, d3, d4 in FIG. 3 are realized by a combination of step 255 and step 260.

Finally, in step 265, command BU (K) is executed according to the contents shown in Table 2 above. This command BU (K) (0 ≦ K ≦ 3) is stored in the form of an array in a predetermined storage area, and each of the commands BU (K) (0 ≦ K ≦ 3) indicating each predetermined process corresponding to the variable K described above. Each command code (stored code) is held. Thereby, for example, when K = 0, the command code BU (0) is selected, and the operation of turning on the semiconductor switching element SWUL is executed via the gate driver 106.
This process corresponds to the off operations e1, e2, e3, and e4 in FIG.

  According to the above control procedure, the above-described logic (gate operation) of gate control can be specifically executed. Further, since the above control procedure is common to each phase, the other phases can be controlled in the same manner as the first phase (U phase). And by these controls, generation | occurrence | production of the big on loss by the high on voltage of the flywheel diode comprised in the conventional system can be suppressed, preventing generation | occurrence | production of the shoot through current resulting from a power supply short circuit state. Further, since the flywheel diode itself is not installed, the above-described conventional reverse recovery current is eliminated, the occurrence of a power supply short-circuit state resulting therefrom is prevented, and the switching loss corresponding to that is suppressed.

  Moreover, if a semiconductor switching element having a high breakdown voltage and a low on-resistance composed of a wide band gap semiconductor is used as a bidirectional semiconductor switching element which is a main component of the inverter, a relatively high power supply voltage of about several hundred volts or more. The inverter can be realized with low loss. Further, the number of output phases can be arbitrarily increased or decreased according to the configuration of the main circuit 101. Further, as mentioned above, these control processes may of course be realized by hardware.

For example, as described above, according to the present invention, in an inverter configured using switching elements capable of bidirectional current control, one switching element of the inverter arm is in a conductive state without parallel connection of flywheel diodes. By switching the other switching element from the non-conducting state to the conducting state without delay when switching from the non-conducting state to the non-conducting state, a regenerative current can be passed through the switching element. According to such a control method, it is possible to achieve both the prevention of the occurrence of a short-through current and the suppression of the time differential value of the motor current.
In addition, since the present invention makes it possible to omit flywheel diodes, the number of expensive main circuit elements can be halved in the inverter according to the present invention, thereby significantly reducing the inverter cost. Can do.

[Other variations]
The embodiment of the present invention is not limited to the above-described embodiment, and other modifications as exemplified below may be made. Even with such modifications and applications, the effects of the present invention can be obtained based on the functions of the present invention.
(Modification 1)
For example, in the first embodiment, each semiconductor switching element used in the main circuit has no parasitic diode, but the semiconductor switching element used has substantially the same function as an intentionally provided normal flywheel diode. In some cases, the floating diode that performs the above is parasitic even if not intended. However, the present invention cannot accept the parasitic of such a floating diode, and an inverter using such a semiconductor switching element has substantially the same operation and effect as the first embodiment based on the present invention. Can be obtained.
However, at least from the viewpoint of reducing switching loss by eliminating reverse recovery current and reducing the on-loss associated with regenerative current, such a floating diode is parasitic on each semiconductor switching element as much as possible. It is desirable not to.

In the inverter of the present invention, a bidirectional semiconductor switching element is used for the main circuit, but the loss reduction technique of the present invention is an effective technique regardless of the level of the power supply voltage. That is, the present invention should not be limited to an inverter having a relatively low power supply voltage of about several tens of volts, and the present invention can be applied to a relatively high power supply of about several hundred volts or more from such a low power supply voltage inverter. It is also suitable for reducing the loss of voltage inverters.
Therefore, the present invention can greatly contribute to a wide range of industries using inverter-controlled motors.

Circuit diagram of the inverter 100 according to the first embodiment. Control block diagram of inverter 100 of Embodiment 1 Control block diagram of dead time management unit 105 Flowchart illustrating the control procedure of dead time management unit 105

Explanation of symbols

DESCRIPTION OF SYMBOLS 100: Inverter 101: Main circuit 102: Each phase midpoint voltage measurement part 103: Each phase current measurement part 104: Each phase midpoint voltage high / low command signal generation part 105: Dead time management part 106: Gate driver

Claims (4)

A pair of semiconductors having a bidirectional semiconductor switching element capable of flowing a current bidirectionally in an ON state in a controlled current path to be controlled for opening and closing, wherein the controlled current paths are connected in series In an inverter provided with a switching element for each phase,
A midpoint voltage measuring means for measuring a voltage at a connection point of the pair of semiconductor switching elements for each phase;
Current detection means for detecting the current direction of each phase;
Dead time management means for adaptively controlling, based on the direction of the current and the voltage, for each phase, the length of the dead time in which both of the pair of semiconductor switching elements are simultaneously turned off,
The dead time management means includes
The increase or decrease in the voltage from the time when the OFF command signal for switching the pair of one of the semiconductor switching elements from the ON state to the OFF state is output indicates a displacement amount exceeding a predetermined threshold (> 0). An inverter that outputs an ON command signal for switching the pair of other semiconductor switching elements from an OFF state to an ON state. The dead time management means includes
1st voltage raising / lowering determination means which determines that the said ON command signal should be output when the said displacement amount reaches the value of 10 to 20% of the direct-current power supply voltage to be used, It is characterized by the above-mentioned. The inverter according to claim 1. The dead time management means includes
A second voltage increase / decrease determination unit that determines that the ON command signal should be output when the displacement amount reaches a value that is twice or more the voltage drop amount when the semiconductor switching element is conductive; The inverter according to claim 1 or 2, wherein the inverter is characterized. The dead time management means includes
The inverter according to any one of claims 1 to 3, further comprising period measuring means for defining an upper limit value of the length of the dead time.

Images