JPH0698583A - Inverter - Google Patents

Abstract

PURPOSE:To avoid the decline of a motor efficiency caused by the delay of the rise of a winding current from a commutation timing in an inverter which facilitates that, in order to drive a brushless motor with PWM(pulse width modulation) control, the rotor rotating position of the motor is recognized by waveform synthesis from the terminal voltage of the motor winding and a recognition waveform having commutation timing information is formed and the commutation timing of the winding is obtained in accordance with the recognition waveform. CONSTITUTION:A current applied to a three-phase bridge circuit 13 is detected by a shunt resistor 32. A time corresponding to an electrical angle of 30 degrees from the rise time of a recognition waveform is used as a reference timing. The reference timing is corrected so as to have a detected current minimum and the commutation timing of a motor winding is determined.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a switching circuit for sequentially energizing each winding of a motor having a plurality of windings, such as a brushless DC motor, at commutation timing corresponding to a predetermined rotation position of a rotor. In particular, the present invention relates to an inverter device having a switching element driven by a pulse width modulation signal.

[0002]

2. Description of the Related Art In recent years, in air conditioners and refrigerators, a brushless motor, which is a kind of DC motor, has been adopted and driven by an inverter device in order to change the capacity of the compressor and save power consumption. In the case of brushless motors, the rotational position signal of the rotor is usually required to determine the energized phase of the winding.However, depending on the environment in which the motor is used, such as when the motor is exposed to refrigerant, as in compressors of air conditioners and refrigerators. It may be difficult to arrange the position detection sensor. For this reason, the inventors of the present application have developed a technique for detecting the induced voltage in the winding of the motor and electrically processing the induced voltage to obtain a rotational position signal, and applied for it as Japanese Patent Application No. 62-162654.

Hereinafter, the case where the invention of this application is implemented by a pulse width modulation (hereinafter, simply referred to as PWM) system will be described as an example, and will be described as a conventional technique with reference to FIGS.

In the inverter device shown in FIG. 5,
The DC power supply circuit 2 connected to the AC power supply 1 comprises a full-wave rectifier circuit 3, a reactor 4a and a smoothing capacitor 4b. Between the DC busbars 5 and 6 of the DC power supply circuit 2, a switching element such as a switching element is provided for switching. Three-phase bridge circuit 13 including transistors 7 to 12
Is connected, and the windings 15u, 15v, 15w of the brushless motor 15 are connected to the output terminals 14u, 14v, 14w thereof.

Each transistor 7 of the three-phase bridge circuit 13
When the .about.12 are on / off controlled in a predetermined order, the brushless motor 15 is rotationally driven by sequentially and repeatedly energizing the windings 15u to 15w thereof with a phase difference of 120 degrees (electrical angle, the same applies hereinafter). In this case, one transistor is controlled by an on / off cycle of 120 degrees on and 240 degrees off, and in the on cycle, the PWM signal P shown in FIG.
Since the duty is controlled by 1, the terminal voltage Vu of each winding 15u to 15w of the brushless motor 15
Vv and Vw have the waveforms shown in FIG.

FIG. 6 shows the waveforms of the terminal voltage Vu and the current Iu of the winding 15u when the PWM control is not involved. In this waveform, the sloped portion over a section of about 60 degrees (period Ta) is the induced voltage of the winding, the elongated positive and negative pulses are the pulse voltage by the diode D connected in parallel with each transistor of the three-phase bridge circuit 13, and V0 is DC bus 5,
6 is a reference voltage formed by the resistance voltage dividing circuit 16 connected between the six. From FIG. 7, it is understood that the commutation timing is delayed by about 30 degrees from the time when the induced voltage crosses the reference voltage V0 (hereinafter referred to as the zero crossing time).

The terminal voltages Vu, Vv, and Vw are compared with the reference voltage V0 by comparators 18 to 20 provided in the energization signal circuit 17, so that the terminal voltages Vu to Vw as shown in FIG. It is converted into fundamental wave signals Vu ′, Vv ′, Vw ′ for section recognition. Further, these fundamental wave signals Vu ′, Vv ′, Vw ′ are supplied to the energization signal circuit 1
7 is applied to the waveform synthesis circuit 21 provided in FIG.
By collating with the WM signal P1, it is converted into recognition waveform signals Ua, Va and Wa which are continuous square waves having a time width of 180 degrees and only having a positive pulse component and having a phase difference of 120 degrees from each other. The starting point (rising point) and ending point (falling point) of the recognition waveform signals Ua, Va, Wa coincide with the crossing point of the induced voltage and the reference voltage V0.

Further, in the waveform synthesizing circuit 21, of the first and second timer functions held therein, the first
The three recognition waveform signals Ua,
Six first phase division patterns X1 to X6 each having a time width Tb of 60 degrees are formed from Va and Wa, and the first phase division patterns X1 to X6 are further formed by the second timer function.
Six second phase division patterns Y1 to Y6 each having a time width of 30 degrees starting from the end point of X6 are formed. Then, the waveform synthesizing circuit 21 finally conducts the energization signals Up, Un, V shown in FIG. 7 from the second phase division signal as described above.
p, Vn, Wp, and Wn are combined.

Here, the energization signals Up, Un, Vp, V
The starting points of n, Wp, and Wn are the second phase division pattern Y.
Since it coincides with the end points of 1 to Y6, it is a time point delayed by 30 degrees from the time point at which the induced voltage crosses the reference voltage V0. Therefore, these energization signals Up, Un, Vp, Vn,
The phase patterns of Wp and Wn coincide with the commutation timing patterns required for the transistors 7 to 12 of the three-phase bridge circuit 13.

On the other hand, the speed judgment circuit 22 judges the speed deviation from the energization signal Wn and the speed command signal Sc given as the rotation speed detection signal of the brushless motor 15 from the waveform synthesis circuit 21, and corresponds to the speed deviation. The velocity deviation signal Sd is output and given to the pulse width modulation circuit 23.
The pulse width modulation circuit 23 controls the duty of the PWM signal P1 according to the magnitude of the speed deviation signal Sd. The PWM signal P1 whose duty is controlled in this way
Is each gate portion 2 of the gate circuit 24 that constitutes the drive circuit.
5, 27, and 29, the energizing signals Up, Vp, W
It is supplied as a base control signal to the bases of the transistors 7, 9 and 11 of the three-phase bridge circuit 13 while being combined with p, for example, and being ANDed. As a result, the transistor 7
To 12 are controlled to be turned on and off in the pattern shown in FIG. 7 by the energization signals Up to Wn, so that the brushless motor 1
5 continues to drive and the PWM signal P shown in FIG.
The speed control is performed by the duty control by 1.

[0011]

As is apparent from the above description, the windings 15u, 15v, 15w of each phase have a delay of 120 degrees, which is delayed by 30 degrees from the time when the induced voltage crosses the reference voltage V0. The section is energized.

8 (a), (b) and (c),
Induction voltage of the U-phase winding 15u of the brushless motor 15 and P
The relationship between the applied voltage and the current without WM control is shown. As is clear from the figure, the voltage of the DC power supply circuit 2 applied to the winding 15u is the peak time Tp of the induced voltage.
A symmetrical waveform is formed in the range of 120 degrees with respect to. On the other hand, the current Iu flowing through the winding 15u rises in an inclined manner from the time when the voltage application of the DC power supply circuit 2 is started, and the time I
It reaches the peak with a delay, and then falls to the inclined state from the point of time when the voltage application is completed and becomes zero with a delay of time T2 (equivalent to T1). Therefore, the current Iu flowing through the winding 15u
The waveform of is asymmetric with respect to the peak time Tp of the induced voltage. This also occurs when the PWM control is performed.

Since the torque generated by the motor is generally represented by the product of the induced voltage and the current, the efficiency of the motor is reduced in the conventional case where the current has a waveform asymmetric with respect to the peak time Tp of the induced voltage. Particularly in air conditioners and the like, rapid cooling and heating are required by maximizing output with a limited power supply capacity, and efficiency improvement is also desired from the viewpoint of energy saving and running cost.

Therefore, as shown in FIG. 8D, the induced voltage crosses the reference voltage V0, not the time when the commutation timing is delayed by 30 degrees from the time when the induced voltage crosses the reference voltage V0. It is conceivable to set it at a time point earlier than the time point delayed by 30 degrees by a certain angle (corresponding to the time Td shown in (d) of FIG. 8). By doing so, the waveform of the current Iu becomes symmetrical with respect to the peak time Tp of the induced voltage, as shown in (e) of FIG. 8, and as a result, the so-called power factor is improved and the current Iu also decreases. .

However, the delay times T1 and T
2 is not always constant, and becomes longer as the load torque (current value) becomes larger, and becomes shorter as the rotation speed (induced voltage) becomes larger. For this reason, even if the commutation timing is set to a certain time point that is delayed by a certain angle smaller than 30 degrees from the time point when the induced voltage crosses the reference voltage V0, a sufficient improvement in efficiency cannot be expected.

The object of the present invention is different in the case where the delay of the rising and falling of the winding current with respect to the start time and the end time of the voltage application to the winding of the motor is not constant due to the load torque and the rotation speed of the motor. Even in such a situation, it is an object of the present invention to provide an inverter device that allows a winding current to flow in a symmetrical waveform with respect to a peak time of an induced voltage and improves the efficiency of a motor.

[0017]

An inverter device according to the present invention is a pulse width modulation circuit for obtaining a pulse width modulation signal, and a switching circuit comprising a plurality of switching elements for sequentially energizing windings of a plurality of phases included in a motor. A quantity of electricity detecting means having a current detecting element for detecting at least a current supplied to the switching circuit, a commutation timing, a comparison result of the terminal voltage of the winding and a reference voltage,
And an energization signal forming unit for determining an energization signal corresponding to the commutation timing, which is determined by a result of successive comparison of the amount of electricity detected by the electricity amount detection unit, and the energization signal and the pulse width modulation signal are combined. And a drive circuit for driving the switching element.

[0018]

According to the present invention, the terminal voltage of the winding of the motor is detected during rotation of the motor, and the quantity of electricity (at least current) supplied to the switching circuit is detected by the quantity of electricity detecting means. Then, for example, first, the terminal voltage is compared with the reference voltage, and based on the result, a fixed timing corresponding to the rotational position of the rotor is obtained, and this fixed timing is determined by successive comparison of the electric quantities supplied to the switching circuit. The commutation timing is obtained by correcting the electricity amount to the smaller side, and the energization signal corresponding to this commutation timing is formed. Then, the switching element is on / off controlled by a composite signal of the energization signal and the pulse width modulation signal, for example, a logical product signal. In this case, the commutation timing is determined so as to reduce the amount of electricity supplied to the switching circuit, so that the efficiency of the motor is increased.

[0019]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT An embodiment of the present invention will be described below with reference to FIGS.
The description will be given with reference to FIG. 5, but only the portions different from FIG. In this embodiment, the waveform shaping circuit 21 in the communication signal circuit 17 in the conventional inverter device shown in FIG. 5 comprises a microcomputer 31. This microcomputer 31 has a waveform shaping circuit 2
In addition to all the functions of No. 1, it has a function of changing commutation timing and a third timer function for recognizing a specific period, as described later.

The recognition of this specific period is based on the recognition waveform signal U.
a, Va, Wa near the rising and falling timings, in other words, each winding 15u of the brushless motor 15,
Terminal voltages Vu, Vv, including induced voltages of 15v, 15w,
This is for recognizing the vicinity of the time when Vw and the reference voltage V0 cross. Therefore, as shown in FIG. 2, the third timer function of the microcomputer 31 has a commutation timing correction time Tc (described later) at a time corresponding to 15 degrees from the end point of each of the second phase division patterns Y1 to Y6. ) Is added and the time Z1 to Z6 is measured.

Here, when the time of the first phase division patterns X1 to X6 is Tb, the time corresponding to 15 degrees is (Tb /
4) It's time. Time Z by this third timer function
The microcomputer 31 is measured by measuring 1 to Z6.
Recognizes a specific period Ti corresponding to a time width of 15 degrees (Tb / 4) having the same end point as the end point (crossing point of the induced voltage and the reference voltage V0) of each of the first phase division patterns X1 to X6. , Within the specific period Ti, the comparator 18
The fundamental wave signals Vu ′, Vv ′, and Vw ′ from ˜20 are input. In this way, by limiting the input of the fundamental wave signals Vu ', Vv', Vw 'to the specific period Ti, the induced voltage and the reference voltage V0 cross at the time when the pulse voltage of the diode D crosses the reference voltage V0. It prevents false recognition of the time.

Here, the recognition by the microcomputer 31 at the time when the induced voltage crosses the reference voltage V0 is as follows.
Fundamental wave signals Vu ′, Vv from the comparators 18 to 20
, Vw 'and the PWM signal P1 from the pulse width modulation circuit 23 and the comparison data stored therein are compared. The comparison data is, as shown in Table 1 below, the fundamental wave signal Vu for each of the first phase division patterns X1 to X6.
', Vv', Vw 'and PWM signal P1 high (H)
-It is configured as low (L) mode.

[0023]

[Table 1] Then, the microcomputer 31 inputs the comparison data corresponding to the next phase division pattern of the phase division pattern currently being executed in each of the first phase division patterns X1 to X6, and outputs the fundamental wave signal within the specific period Ti. Vu ',
When the high / low states of Vv ', Vw' and the PWM signal P1 coincide with the input comparison data of the phase section, it is recognized that the induced voltage crosses the reference voltage V0.

Further, in this embodiment, the speed judging circuit 22 in the conventional inverter device shown in FIG. 5 is omitted, and the function of detecting the rotation speed of the motor and the PWM signal P1 of the speed judging circuit 22 are omitted. The microcomputer 31 bears the function of determining the duty. That is, in the first phase division patterns X1 to X6, the microcomputer 31 determines the time of the motor from the sum of the time of 6 patterns (motor half rotation) or 12 patterns (motor one rotation) before the phase division pattern currently being executed. The number of rotations per unit time (rotational speed: hereinafter, simply referred to as the number of rotations) is determined, and this is compared with the number of rotations obtained from the speed command signal Sc given from the outside to determine the speed deviation. Corresponding duty signal Sd
To the pulse width modulation circuit 23. Then, the pulse width modulation circuit 23 outputs the PWM signal P1 having the duty D1 shown in the duty signal Sd.

In this embodiment, the current is detected from the amount of electricity supplied to the three-phase bridge circuit 13, which is a switching circuit, and the commutation timing is changed so that the detected current is minimized. It A shunt resistor 32 (current detection element) that generates a voltage corresponding to the current is connected to the DC bus 6 in order to detect the current flowing in the three-phase bridge circuit 13 as the electricity amount detecting means. The detection voltage of the resistor 32 is input to the microcomputer 31. The microcomputer 31 includes an A / D converter, digitizes the detection voltage of the shunt resistor 32, and recognizes it as a current value.

Next, the operation of the above configuration will be described. First, in this embodiment, the reference timing is a time delayed by 30 degrees from the time at which the induced voltage crosses the reference voltage V0, while the current detected from the detection voltage of the shunt resistor 32 is sequentially compared, The correction time Tc is set so that the current becomes smaller, and the commutation timing is set to be earlier than the reference timing by the correction time Tc.

This will be specifically described with reference to the flow charts shown in FIGS. Note that FIG.
The routines (a) to (c) are configured as interrupt routines with respect to the routine shown in FIG.

First, FIG. 4A is a routine for determining the duty for speed control, and this routine is performed once for each revolution of the brushless motor 15, for example. When this duty determination routine is executed, first, the command rotational speed Nc indicated by the speed command signal Sc is input (step S1), then the actual rotational speed N of the motor 15 is determined (step S2), and then the duty D1 Is calculated by the following equation, and the result is the duty signal Sd
Is output to the pulse width modulation circuit 23 (step S3). D1 = D0-A * (N-Nc) A is a gain constant.

Further, FIG. 4B shows a three-phase bridge circuit 13
It is a routine for detecting the current flowing through the device and is performed every 100 microseconds. When this routine is executed,
First, the detection voltage of the shunt resistor 32 is input, the current value corresponding to this detection voltage is obtained, and the current value is integrated (step SI). In the next step SII, it is judged whether or not one second has elapsed from the start of current value integration, and if it has not yet elapsed, the routine returns. Then, when 1 second has elapsed from the start of current value integration (“YES” in step SII), the average current I is calculated from the current integrated value, and the current integrated value is cleared (step SIII), and the process returns.

Now, FIG. 4C is a routine for determining the correction time Tc, which is performed at a cycle longer than the execution cycle of the current detection routine. In this routine,
First, in step SA, it is determined whether or not the brushless motor 15 is in a stable state in which there is no fluctuation in rotation speed. In this determination, since the rotation speed variation involves a change in duty, the amount of change in duty is monitored, and when the amount of change is equal to or less than a predetermined value, it is determined to be in a stable state.

When the state is unstable in step SA, the rotational speed fluctuates due to load torque fluctuations, voltage fluctuations, etc., and the duty varies accordingly. In this state, the three-phase bridge circuit Since the current flowing through 13 is fluctuating, it is not appropriate to carry out the current comparison in the next step SB.
It becomes "O" and returns. When it is determined in step SA that the current state is stable, in the next step SB, the current I obtained in step SIII when the latest current detection routine is executed and the current I obtained in step SIII one second before that.
Compare with ´.

As a result of the comparison in this step SB, I '<
If I, the process proceeds to the next step SC, and it is determined whether or not the correction time Tc was increased at the time of executing the correction time determination routine of the previous time. The correction time T at the time of the last execution of this routine
If c has been increased, the process proceeds from step SB to step SF, in which the correction time Tc is set to the time obtained by subtracting the increment / decrement unit time A from the previous correction time Tc, and the process returns. On the contrary, the correction time Tc is decreased at the last execution of this routine (in this embodiment, I = I at step SB).
(Including the case where the return is'),
The process proceeds from step SB to step SD, where the correction time Tc is set to a time obtained by adding the increase / decrease unit time A to the previous correction time Tc, and the process returns.

The reason why the increase / decrease in the correction time Tc is opposite to the increase / decrease in the previous correction time Tc is that I '<I due to the increase or decrease in the previous correction time Tc (three-phase bridge circuit 13). Since the current flowing in the current has increased), if the correction time Tc was previously increased in order to decrease the current, the current time is decreased conversely, and if the correction time Tc was previously decreased, the current time is reversed. I am trying to increase it.

On the other hand, as a result of the comparison in step SB, I '
If> I, the process proceeds to step SE, and it is determined whether or not the correction time Tc was increased during the previous execution of the correction time determination routine. If the correction time Tc was increased during the previous execution, the current has decreased due to the increase in the correction time Tc. Therefore, the correction time Tc is also increased this time, and thus the current is decreased. Step SD is executed, where the correction time Tc is set to the time obtained by adding the increment / decrement unit time A to the previous correction time Tc and the routine returns. On the contrary, if the correction time Tc was decreased at the time of executing this routine the previous time, the current has decreased due to the decrease in the correction time Tc, and therefore the process proceeds to step SF which is the side promoting the current decrease. Then, the correction time Tc is
It is set to a time obtained by adding the increment / decrement unit time A to the previous correction time Tc and the process returns.

Now, in the main routine shown in FIG. 3, assuming that the specific period Ti of a certain phase division pattern among the first phase division patterns is entered, the next phase division pattern is compared in step ST1. Data is loaded and the fundamental wave signal Vu input during the specific period Ti is input.
′, Vv ′, Vw ′ and the high / low states of the PWM signal P1 are compared with the comparison data (step ST2). When the induced voltage crosses the reference voltage V0, the high-low states of the fundamental wave signals Vu ', Vv', Vw 'and the PWM signal P1 match the comparison data ("YES" in step ST2). It is the start of the next first phase division pattern, and in step ST3 the time required for the immediately preceding first phase division pattern Tb is loaded, and at the same time the time required for the started first phase division pattern is measured. Restart the timer function of 1.

Then, in the next step ST4, the time of the second phase division pattern is calculated. In this case, assuming that the time of the second phase division pattern is (Tb / 2), the commutation timing is a point delayed by 30 degrees from the cross point of the induced voltage and the reference voltage V0 (the same point as the conventional one). Therefore, this is corrected by the equation [(Tb / 2) -Tc] using Tc set in the correction time determination routine, and the second timer function is started. As a result, the commutation timing is corrected to a time point Tc earlier than the time point delayed by 30 degrees from the crossing point of the induced voltage and the reference voltage V0.

Next, the number of sections of the first phase division pattern is incremented (step ST5), and when the second timer function finishes counting [(Tb / 2) -Tc] ("YES" in step ST6). ), The next step ST7
The energization signal is output with. Then, in the next step ST8, the measurement time of the third timer function is calculated. Here, if the measurement time of the third timer function is set to ¼ of the time Tb of the first phase division pattern, the time of the second phase division pattern is earlier than (Tb / 4) by Tc. However, the specific period Ti starts from a point earlier than Tc. Therefore, in the present embodiment, the measurement time of the third timer function is obtained by the formula [(T / 4) + Tc], and the specific period Ti is approximately 15 degrees before the crossing point of the induced voltage and the reference voltage V0. I'm trying to get started. When the third timer function finishes counting the time and enters the specific period Ti (“YES” in step ST9), the comparison data of the next first phase division pattern incremented in step ST5 is loaded. Step ST
Return to 1.

According to this embodiment, the commutation timing is advanced from the conventional timing by the correction time Tc. The correction time Tc is determined so that the current I flowing through the three-phase bridge circuit 13 is minimized, as understood from the correction time determination routine of FIG. Therefore, the efficiency of the brushless motor 15 is improved, and the efficiency is improved by controlling the current to the minimum, which means that the so-called power factor is improved, and the windings 15u, 15v, 15
This means that the current waveform of w is symmetrical with respect to the peak time Tp of the induced voltage.

In the above embodiment, the reference timing is a point delayed by 30 degrees from the point at which the induced voltage and the reference voltage V0 cross, and this is corrected by the correction time Tc to obtain the commutation timing. Induction voltage and reference voltage V0
It is also possible to use the time when the points cross as the reference timing and correct this with the correction time (delay time from the reference timing) to obtain the commutation timing. Further, in the above-described embodiment, the electric current is detected from among the electric quantities supplied to the three-phase bridge circuit 13, and the correction time Tc is calculated based on the detected electric quantity.
However, the current and voltage may be detected, and the correction time may be obtained from both of them.

[0040]

As described above, according to the present invention, even if the rotation speed and the load of the motor change, the commutation is such that the winding current always has a symmetrical waveform with respect to the peak time of the induced voltage of the motor winding. It is possible to provide an inverter device capable of ensuring timing and, as a result, improving motor efficiency.

[Brief description of drawings]

FIG. 1 is a circuit diagram showing an embodiment of the present invention.

FIG. 2 is a waveform diagram of each part in FIG.

3 is a flowchart for explaining the operation of the microcomputer shown in FIG.

FIG. 4 is a flowchart 2

FIG. 5 is a circuit diagram of a conventional inverter device.

FIG. 6 is a waveform diagram of the terminal voltage and current of one winding of the brushless motor.

FIG. 7 is a view corresponding to FIG.

FIG. 8 is a waveform diagram of induced voltage, terminal voltage, and current of one winding of a brushless motor.

[Explanation of symbols]

2 is a DC power supply circuit, 7 to 12 are transistors (switching elements), 13 is a three-phase bridge circuit (switching circuit), 15 is a brushless motor, 17 is an energization signal circuit,
Reference numeral 23 is a pulse width modulation circuit, 24 is a gate circuit (driving circuit), 31 is a microcomputer (energization signal forming means), and 32 is a shunt resistor (electric quantity detecting means, current detecting element).

─────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] January 6, 1993

[Procedure Amendment 1]

[Document name to be amended] Statement

[Correction target item name] 0039

[Correction method] Change

[Correction content]

In the above embodiment, the reference timing is a point delayed by 30 degrees from the point at which the induced voltage and the reference voltage V0 cross, and this is corrected by the correction time Tc to obtain the commutation timing. Induction voltage and reference voltage V0
It is also possible to use the time when the points cross as the reference timing and correct this with the correction time (delay time from the reference timing) to obtain the commutation timing. Also, the above
In the embodiment, the current flowing through the DC bus 6 is changed to the three-phase bridge circuit.
Although it is detected as the current supplied to the path 13,
Flow from AC power supply 1 to full-wave rectifier 3 of DC power supply circuit 2
As the current supplied to the three-phase bridge circuit 13,
It may be configured to detect. Furthermore, in the above embodiment, among the electric quantities supplied to the three-phase bridge circuit 13,
Although the current is detected and the correction time Tc is obtained based on the amount of the current, the current and voltage may be detected and the correction time may be obtained from both of them.

Claims (1)

[Claims] 1. A switching circuit including a plurality of switching elements for sequentially energizing windings of a plurality of phases included in a motor, a pulse width modulation circuit for obtaining a pulse width modulation signal, and at least a current supplied to the switching circuit. The amount of electricity detecting means having a current detecting element for detecting the commutation timing, the result of comparison of the commutation timing between the terminal voltage of the winding and the reference voltage, and the amount of electricity sequentially detected by the amount of electricity detecting means. And a drive circuit for synthesizing the energization signal and the pulse width modulation signal to drive the switching element.