DE102007057746A1 - Lathe i.e. fuel pump actuator, driving device, has control unit for controlling lathe, and drive control device for stopping drive of lathe when indication is detected, to bring lathe to freely running condition - Google Patents

Abstract

The device (1) has a synchronization loss projection device (7) for monitoring a condition of a turn of a lathe i.e. fuel pump actuator, for detecting an indication of the lathe that switches into a condition of a synchronization loss. A drive control device (4) stops a drive of the lathe when the indication is detected, to bring the lathe to a freely running condition. A control unit controls the lathe. The projection device detects a number of revolutions of the lathe, and compares detected number of revolutions with a normal number of revolutions of the lathe.

Description

Field of the invention

The The present invention relates to an apparatus and a method for driving electric lathes or rotating Machines or rotary machines, such as a brushless DC motor, the rotor position the lathe appreciated becomes the energization timing for driving the lathe determine.

Background of the invention

Some conventional Actuators use a sensorless position method which is adapted to the rotor position of a brushless DC motor estimate and thereby achieve the commutation of the engine and to drive him. If a fault occurs in such a drive device or a load fluctuation it may become a state of loss of synchronization be brought in which he can no longer control the engine, as it is thought.

The JP-A-2004-104935 discloses a technology for resuming control of driving with respect to the motor when it is detected that a motor has been brought into a state of synchronization loss and has been stopped. However, according to this method, in a case of an engine or the like for driving an electric vehicle, it is not appropriate to stop the rotation of the engine while the vehicle is running even if it has been brought to a state of synchronization loss. The rotation of the engine must be maintained as much as possible. In this method, after it has been detected that a motor has completely lost synchronization, the synchronization loss is coped with.

The JP 4-317587A , the US 5,432,414 ( JP 5-284781A ) and the JP 7-327390A disclose technologies for starting the engine by changing an excitation frequency (excitation) of the motor. Such technologies are proposed because a motor is temporarily rotated in a reverse direction or too large a torque generated in a motor, and this causes too high a speed and a loss of synchronization, resulting in a prolonged start time in the conventional device. However, the proposed circuit for changing an excitation frequency is complicated and inevitably increases the size of the circuit. For example, when an excitation frequency is changed by digital processing, the number of bits of a counter for counting cycles corresponding to one frequency is increased.

The JP 11-18478A discloses a technology for detecting a timing at which an electric angle of a motor becomes equal to a predetermined electrical angle based on an induced voltage developed as a terminal voltage of the motor. According to this technology, a restriction is imposed on a permitted duration for which detection of a time is released when the predetermined electrical angle occurs. However, a detected value of the rotational speed of a motor transitions to an excessively high value or too low a value and is set there. In these cases, it is difficult to control the rotational state of the engine as desired. When it goes too high or too low, a time when a predetermined electrical angle occurs does not fall within the released period. There are cases where, for example, a power supply voltage or the load on a motor suddenly fluctuates, causing the engine speed to suddenly fluctuate. Also, in these cases, a time when a predetermined electrical angle temporarily falls outside the released period may occur. For this reason, when a time when a predetermined electric angle does not occur within a released period, the rotation state is determined to be abnormal. There is a possibility that both a state (state of synchronization loss) in which it is difficult to control the rotational state as desired and a state of temporary rotational fluctuation caused by a load change or the like are determined to be abnormal. It is therefore difficult, for example, to continuously control a lathe when only a load change occurs.

The US 2005/0258788 ( JP 2005-333689A ) discloses determining an electrical angle of a motor by detecting induced voltages, that is, terminal voltages. When a three-phase motor is started, all the switching elements of an inverter are turned off, and therefore each phase of the three-phase motor is in a high-impedance state. For this reason, a situation in which a zero-point voltage is equal to the potential of each phase of the three-phase motor may occur. When noise is mixed when the induced voltage is detected in this state, the zero crosses point voltage and the voltage of each phase frequently. Finally, the zero crossing time is often detected incorrectly. For this reason, for example, a system required to operate an inverter based on a detection signal with respect to a zero-crossing time immediately after a start of a three-phase motor can not suitably satisfy this requirement.

Furthermore, in the US 2005/0258788 the time required for the rotor to rotate by a predetermined interval of electrical angle is determined from time intervals between occurrences of a time at which the preceding zero crossing occurs. A time at which the required time passes through an occurrence of a zero-crossing timing is taken as a certain time point at which an angle occurs that provides a basis for a switching operation. When a certain time is set by the foregoing method when the three-phase motor is started, the specific time is set by shortening the predetermined interval of the electrical angle used in the preceding calculation of the required time. If this time is calculated in the initial stage of start-up as under normal conditions, this time is invariably delayed from a time point at which a reference angle occurs. In this case, the predetermined timing is set by determining a required time from an occurrence of a zero-crossing timing to when a reference angle occurs based on a time interval between occurrences of the zero-crossing timing. The inventors have found that in order to perform this determination with accuracy, the speed of the motor must be stable. For this reason, the timing at which the reference angle occurs can be set not only accurately when the engine is started, but also generally when the rotational speed fluctuates greatly. This can lead to a deteriorated controllability of the engine.

The JP 2642357B1 discloses an example of a conventional control apparatus for multiphase lathes. In another method of controlling a lathe (a three-phase brushless motor), a 120 ° excitation method is proposed that in 50 is shown. In this figure, (a) represents the transition of terminal voltages Vu, Vv, Vw; (b) illustrates the transition of comparison signals PU, PV, PW as a result of comparing the terminal voltages Vu, Vv, Vw shown by solid lines in (a) with a reference voltage Vref; represents (c) the transitions of a one-bit combination signal PS obtained by logically combining the comparison signals PU, PV, PW; and represents (d) the transition of a detection signal Qs obtained by shaping the waveform of the combination signal PS. At a time (zero-crossing timing) at which the terminal voltages Vu, Vv, Vw shown in (a) coincide with the reference voltage Vref, the output of the comparison signals PU, PV, PW is inverted. Actually, however, the output of the comparison signals PU, PV, PW is also inverted when the operation of the switching elements of an inverter (a power converter circuit) connected to the brushless motor is changed. This inversion is caused by the passage of current through diodes connected in parallel with the switching elements. For this reason, the rising edges and falling edges of the combination signal PS obtained by logically synthesizing the comparison signals PU, PV, PW do not coincide with only one zero-crossing timing. Some of them coincide with a time when a current is supplied through the diodes. Meanwhile, all the rising edges and the falling edges of the detection signal Qs obtained as a result of waveform shaping coincide with a zero-crossing timing.

Of the electrical angle of the brushless Motors is uniquely determined by a zero crossing time. For this reason, the following can be changed by changing the operating state realized by switching elements at the time (the specific time) when a time has elapsed that is required a motor by a predetermined angular interval (for example, 30 °) from the Zero crossing time turns. The brushless Motor can through a 120 ° excitation process to be controlled. More specifically, a time sequence pattern regarding the Operation of switching elements determined. That's why a tax can through the 120 ° excitation process by operating the switching elements according to the preceding pattern be achieved at any time at which the particular time occurs.

Since the detection signal Qs is a one-bit signal, it is impossible to distinguish a zero-crossing time from another in the three-phase brushless motor in accordance with the signal. For this reason, when the rotation state of a brushless motor becomes abnormal or noise in a terminal voltage Vu, Vv, Vw or the like is mixed, there is a possibility that the controllability of the brushless motor is significantly deteriorated. A more detailed description will be given. For example, even when the brushless motor is rotated, it is difficult to detect this reverse rotation from the detection signal Qs. Therefore there is a possibility that a change of operation of the Switching elements as under normal conditions is continued when a time required by a rising edge or a falling edge of the detection signal Qs has elapsed (certain time). In this case, the brushless motor can not be appropriately controlled.

It is a method of execution the following for the purpose of controlling the output of the brushless Motors, to control and limit a current, the brushless Motor is supplied, or for other similar ones Purposes known. While a shared duration for the switch-on of switching elements based on of the previous predetermined time, becomes PWM modulation processing executed to repeatedly turn the switching elements on and off. In this Case, however, a problem arises. In PWM modulation processing Switching elements are often used by an on-state is switched to an off-state and thereby a stream frequently passed through diodes. After all the comparison signals PU, PV, PW and the combination signal PS often inverted. At this time, it is difficult to get the detection signal Qs as a convenient signal which synchronizes at a zero crossing time is. Therefore, it is difficult to set a specific time appropriately.

Summary of the invention

It Therefore, a first object of the present invention is a lathe driving device and to provide a lathe drive method in which a drive-controlling be recovered without stopping the operation of a lathe can, which is driven by a sensorless process, before the lathe completely is brought to a state of synchronization loss.

To the Achieving the first object, a state of rotation of the lathe is monitored to detect an indication that the lathe becomes a condition a loss of synchronization, and a driving of the lathe is temporarily stopped to the Lathe to bring to a freewheeling state, if the indication is detected. Therefore, a normal control for driving the Lathe resumed.

It It is a second object of the present invention to provide a lathe driving apparatus and to provide a lathe driving method in which a Lathe started in a short time by a simple design can be. To accomplish the second task, a compulsory commutation a lathe running and a current supplied to one turn of the lathe becomes limited to an upper limit that is set higher than a level is, on which a current flows, when the lathe is in a normal rotational state, when the forced commutation is executed.

It It is a third object of the present invention to provide a lathe driving device to be able to, a rotation state of a lathe on the basis of an induced voltage of the engine expedient to to capture. To achieve the third task will be a released Duration, for which detects a predetermined electrical angle the basis of a detected value of a terminal voltage of a Lathe is allowed, and a rotation state of the lathe detected as abnormal when the number of times that the predetermined electrical angles either before or after the released duration occurs, equal or higher as a threshold.

It A fourth object of the present invention is a lathe driving device which is capable of erroneously detecting a zero-crossing timing to avoid to which a zero-point voltage equal to a reference voltage becomes. To achieve the fourth object, a terminal voltage a lathe with a reference voltage with respect to a Amplitude compared to detect a zero crossing time, if the reference voltage, which is either a zero point voltage the lathe or equivalent of which is, and an inverted voltage of the lathe coincide with each other. A switching element for feeding a current of the lathe is based on the zero-crossing timing operated. At least one of a value of the terminal voltage, the is to be compared when a speed of the lathe is substantially Is zero, and a value of the reference voltage is offset corrected, to distinguish the values of the terminal voltage and the reference voltage.

It is a fifth object of the present invention to provide a lathe driving apparatus in which information concerning the electrical angle of a lathe based on a comparison result of an induced voltage of the lathe with a reference voltage a higher accuracy can be detected. In order to achieve the fifth object, a terminal voltage of each phase of a lathe is compared with a reference voltage, and information concerning an electrical angle of the lathe is detected based on a comparison result when a zero-crossing timing in a current operating state of switching elements and an actual comparison result occurs with respect to each phase. It can be determined whether there is an abnormality in a rotating state of the rotating machine based on a detected value of an induced voltage of the rotating machine, and all of the phases of the rotating machine can be brought to either the positive or the negative pole of a power supply, and thereby the Rotation of the lathe forcibly stopped when an anomaly is detected.

It A sixth object of the present invention is a lathe driving device when switching elements of an energy converter circuit operated to control a lathe, a point in time to which a reference angle occurs, based on a Zero transit time determined more appropriate to which an induced voltage of the lathe is disregarded a fluctuation of a rotational speed becomes equal to a reference voltage. To accomplish the sixth task is information that is on a change a rotational speed of a lathe, from a detection result of the zero crossing time point and becomes a particular one Time to control the lathe based on the information focusing on the change the speed refers, changeable established. The information is an acceleration. A height of one Excitation of the lathe will be in accordance with the acceleration limited. The lathe comes with a stream of part of phases to another part of phases supplied to this before the lathe is started, so that a rotation angle of the lathe on a predetermined angle is set. At least one of the one Part of phases and the other part of phases includes one Plurality of phases.

Short description of the drawing

The The present invention will be described below with reference to exemplary embodiments explained in more detail with reference to the accompanying drawings.

It shows:

1 a circuit diagram of a lathe driving device according to a first embodiment of the present invention;

2 Fig. 10 is a flowchart of processing executed by a drive control circuit and a synchronization loss monitoring circuit;

3 a signal diagram of output voltage waveforms of an inverter unit, which are observed when a motor rotates;

4 a signal diagram of a second embodiment of the present invention;

5 a signal diagram of a fluctuation of a switching signal when a shift is performed earlier;

6 a signal diagram of a fluctuation of a switching signal when a switching is performed irregularly;

7 a circuit diagram according to a third embodiment of the present invention, the 1 corresponds;

8th a signal diagram that 3 corresponds;

9 a circuit diagram according to a fourth embodiment of the present invention, partially 1 corresponds;

10 an operation diagram of a state in which a current fluctuates before a motor loses synchronization;

11 a circuit diagram of a lathe driving device according to a fifth Ausfüh Example of the present invention;

12 a signal diagram of voltages of respective phases and position signals of a rotor, which are observed when a motor rotates in a normal state;

13 a circuit diagram of a gate drive circuit;

14 FIG. 12 is a waveform diagram of a waveform developed in when a gate driving circuit restricts a starting current; FIG.

15 a signal diagram of voltages of the respective phases and a motor current, which are observed when a current supplied when an engine is actually started is limited to 2A;

16 an explanatory view of a change of a start time, which is observed when a current is not limited and limited to different limit levels when an engine is started;

17 Fig. 10 is a flowchart of a part of processing executed by the drive control circuit when the engine is started;

18 FIG. 10 is a circuit diagram of a brushless motor lathe driving apparatus according to a sixth embodiment of the present invention; FIG.

19 a signal diagram of a mode for performing a 120 ° excitation control;

20 a flowchart of processing for setting counters for a 120 ° excitation control;

21 a flowchart of a mode for operating switching elements;

22 a flowchart of a processing for correcting a Zählgeschwindigkeit of counters;

23 a characteristic diagram of a relationship between a voltage change and a Zählgeschwindigkeits correction coefficient;

24 a flowchart of a processing for determining a state of a synchronization loss;

25 a flowchart of a processing for determining a state of a synchronization loss;

26A a flowchart for correcting the counting speed of counters according to a seventh embodiment of the present invention;

26B a characteristic diagram of a relationship between an acceleration and a Zählgeschwindigkeits correction coefficient;

27A a signal diagram of a method for setting a maximum increase counter according to an eighth embodiment of the present invention;

27B a signal diagram of a count of a maximum increase counter;

28 a flowchart of determining a state of a synchronization loss;

29 FIG. 13 is a table showing a relationship of a count speed correction coefficient with respect to a voltage and a voltage change according to a first embodiment of the present invention; FIG.

30 FIG. 12 is a table showing a relationship of a count speed correction coefficient with respect to a voltage change and an engine temperature according to another embodiment of the present invention; FIG.

31 a circuit diagram of a lathe driving device according to a ninth embodiment of the present invention;

32 a signal diagram of a mode for 120 ° excitation control;

33 a signal diagram of a mode for detecting a reverse rotation;

34 an operational diagram of a relationship between a forward rotation and a reverse rotation;

35 a flowchart of a restart processing at the time of a reverse rotation;

36A to 36E Signal diagrams of a transition of a reference voltage and terminal voltages, which occur when a brushless motor is started, the rotation has been stopped;

37 a circuit diagram of a comparator according to a tenth embodiment of the present invention;

38 a circuit diagram of a lathe driving device according to an eleventh embodiment of the present invention;

39 a signal diagram of a mode for 120 ° excitation control;

40 a flowchart of a processing for operating switching elements;

41 a flowchart of a processing for changing the operation of the switching elements;

42 a flowchart of processing in a PWM control;

43 a signal diagram of a mode for PWM control;

44 a signal diagram of a mode of reverse rotation;

45 an operational diagram of a relationship between a forward rotation and a reverse rotation;

46 FIG. 10 is a flowchart showing processing for restarting a brushless motor according to a twelfth embodiment of the present invention; FIG.

47 a signal diagram of a basis of a wire break according to a thirteenth embodiment of the present invention;

48 a flowchart of a processing for detecting a wire break;

49 a circuit diagram of a lathe driving device according to a fourteenth embodiment of the present invention;

50 a signal diagram of a mode of operation of a conventional 120 ° excitation control;

51 a circuit diagram of a lathe driving device according to a fifteenth embodiment of the present invention;

52 a signal diagram of a mode of a switching control;

53 a signal diagram of a mode of a switching control;

54 a flowchart of a processing for setting various counters;

55 a flowchart of a processing for switching switching elements;

56A and 56B Signal diagrams of a deviation of an induced voltage between a normal and an acceleration time;

57A and 57B Graphs of results of experiments relating to a detection error of the angle of a rotor during acceleration;

58A and 58B a flowchart and a data table of a processing for setting a specific time point;

59 a signal diagram of a simulation result with respect to the manner in which a speed is increased when a motor is started with a correction;

60 a signal diagram of a simulation result with respect to the manner in which a speed is increased when a motor is started without correction;

61A and 61B a flowchart and a data table of a processing for setting a specific time point according to a sixteenth embodiment of the present invention;

62 FIG. 10 is a flowchart of processing for estimating a zero-crossing timing according to a seventeenth embodiment of the present invention; FIG.

63 FIG. 10 is a flowchart of processing for restricting a current according to an eighteenth embodiment of the present invention; FIG.

64A and 64B Circuit diagrams of a positioning processing performed before a brushless motor is started according to a nineteenth embodiment of the present invention;

65A and 65B Signal diagrams of a time it takes for the angle of an engine to be settled by the positioning processing;

66 a graph of a relationship between an angle at which a rotor is determined by the positioning processing, and a start time;

67 a signal diagram of a relationship between an angle at which a rotor is determined by a positioning processing, and a torque that is generated when a startup operation is started;

68 a flowchart of a positioning processing;

69 a flowchart of a positioning processing according to a twentieth embodiment of the present invention;

70 a flowchart of a positioning processing according to a twenty-first embodiment of the present invention;

71 FIG. 10 is a flowchart of positioning processing according to a twenty-second embodiment of the present invention; FIG.

72 FIG. 10 is a flowchart of processing for restricting a level of excitation in positioning processing according to a twenty-third embodiment of the present invention; FIG.

73 FIG. 10 is a flowchart showing processing for starting a startup of a brushless motor according to a twenty-fourth embodiment of the invention; FIG. and

74 a signal diagram of a transition of a voltage of a battery at startup.

Detailed description the embodiments

First embodiment

It will open 1 directed. To a lathe driving device 1 A driving power supply voltage VB is applied from a battery (not shown) for vehicle driving power. A brushless DC or DC motor 2 , which is a lathe, is driven by an inverter unit 3 driven. The inverter unit 3 is as an energy converter circuit by, for example, connecting six N-channel power MOSFETs 3a to 3f built in a three-phase bridge construction. The output connections of the respective phases of the inverter unit 3 are each with the stator coils (windings) 2U . 2V . 2W the respective phases of the engine 2 connected. The arrows pointing down in the figure indicate mass.

The inverter unit 3 is controlled by a drive control circuit 4 controlled, which is composed of a microcomputer or a logic circuit. Drive signals are via gate drive circuits 5a to 5f to the gates of the FETs 3a to 3f created. comparators 6U . 6V . 6W compare the output voltage of each phase of the inverter unit 3 with a virtual zero point potential. Then, they give comparison signals PU, PV, PW to the drive control circuit 4 and a synchronization loss monitoring circuit and a synchronization loss predicting device, respectively 7 out. The terminals (+) of the comparators 6U . 6V . 6W are connected to the output terminals OUTu, OUTv and OUTw of the respective phases of the inverter unit 3 connected. The terminals (-) of the same are common with a reference voltage source 8th which is equivalent to the virtual zero potential (or VB / 2).

The drive control circuit 4 generates a commutation pattern signal for the inverter unit 3 on the basis of the comparison signals PU, PV, PW and are available via the corresponding gate drive circuit 5 to the gate of each FET 3 out. Similar to the drive control circuit 4 is the synchronization loss monitoring circuit 7 built from a microcomputer or a logic circuit. It detects a sign (possibility) of the engine 2 to proceed to a state of synchronization loss based on the previous comparison signals PU, PV, PW. When the indication is detected, there is a drive stop signal only for a predetermined duration to the drive control circuit 4 out. As long as the drive stop signal is output, the drive control circuit stops 4 a drive-controlling the inverter unit 3 to the engine 2 to let in a freewheeling state.

The drive control circuit 4 and the synchronization loss monitoring circuit 7 perform a processing of 2 out. When power to the drive device 1 is turned on, energizes the drive control circuit 4 the turns 2U . 2V . 2W of the motor 2 with a DC current to position its rotor (step 1: S1). After that, the turns become 2U . 2V . 2W is energized according to a predetermined Kommutationsmuster and thereby the engine by a forced commutation is started (S2).

When the number of revolutions of the engine 2 Increased to some extent after it has been started, an induced voltage can be monitored in the windings 2U . 2V . 2W is developed. As a result, the drive control circuit changes 4 the driving method for the motor 2 to a sensorless mode (S3). That is, a commutation pattern signal for the inverter unit 3 is generated on the basis of the comparison signals PU, PV, PW, and gate signals UH to WL become the gates of the individual FETs 3a to 3f output. The zero-crossing point of an induced voltage has a phase delay of an electrical angle of 30 ° from an appropriate energization time. Therefore, the drive control circuit 4 this phase delay when generating a commutation pattern signal.

While the engine 2 in the sensorless mode, the synchronization loss monitoring circuit detects 7 Data for predicting a synchronization loss (S4). On the basis of these data, it checks whether there is an indication (possibility) of a synchronization loss in the drive state of the motor 2 there or not (S5). If it is determined that there is no indication of a synchronization loss ("NO"), the processing advances to S7.

In S7 checks the drive control circuit 4 whether the engine 2 is rotated at this time or not, on the basis of the comparison signals PU, PV, PW. When the engine is rotated ("YES"), the processing returns to S3 and the sensorless mode is continued. If the engine 2 is stopped ("NO"), the processing returns to S1 and becomes the motor 2 by performing the processing of an initial one Positioning and then restarted a forced commutation.

The prediction and detection of a synchronization loss in S4 and S5 is carried out as described below. 3 shows the waveforms of the output voltages of the inverter unit 3 which are observed when the engine is on 2 rotates. When a three-phase motor is driven, its voltage waveforms are as follows. Two phases energized between a high side and a low side are brought to a high level and a low level, and the remaining phase which is not energized is in a high impedance state. During this period, induced voltages Vu, Vv, Vw develop in the turns 2U . 2V and indicate a transient voltage change between a high level and a low level. In 3 For example, the induced voltages Vu, Vv, Vw are shown as increasing or decreasing linearly during the previous period. In fact, however, the induced voltages change sinusoidally.

The intervals (between different phases) at which the zero crossing point of an induced voltage for a non-energizing period is generated in each phase are a duration T60 equivalent to an electrical angle of 60 °. When a commutation pattern is changed, as it flows in 3 , (a) to (c), is a current back to the freewheeling diodes of the FETs 3a to 3f for an initial duration and a "zero crossing" point is generated. Therefore, the previous duration is reflected in the comparison signals PU to PW, that of the comparators 6U to 6W be issued. However, this duration becomes from waveform processing in the drive control circuit 4 and the synchronization loss monitoring circuit 7 not considered. As a result, position signals PU ', PV', PW 'are generated as shown in FIG 3 , (d) to (f) is shown.

The synchronization loss monitoring circuit 7 detects the zero-crossing interval duration T60 between phases based on the position signals PU ', PV', PW '. Then it checks if the duration T60 equals the time that the normal number of revolutions of the engine 2 corresponds, is or is not. For example, if the normal number of revolutions of the engine 2 10000 rpm at a rated speed and the number of poles of the motor 2 N is 1/6 of the rotation time per unit time (2 / N) ms. Therefore, the synchronization loss monitoring circuit determines 7 when the zero-crossing interval duration T60 becomes longer than (2 / N) ms by a predetermined time that the motor 2 probably goes to a state of synchronization loss ("YES" in S5). Then, it gives a drive stop signal to the drive control circuit 4 out.

Then the drive control circuit stops 4 a drive-controlling with respect to the motor 2 for the duration for which the drive stop signal has been issued. Therefore, she keeps the engine 2 in a freewheeling state (freewheeling control, S6). The time for which the engine 2 is kept in a freewheeling state at this time, for example, several hundred μs to several ms, etc. After the engine 2 is held in a coasting state only for the predetermined period, the drive control circuit proceeds 4 to S7. When the rotation of the engine 2 has not been stopped ("YES"), it continues the drive control in the sensorless mode.

According to the first embodiment of the present invention, the following is realized. The synchronization loss monitoring circuit 7 the lathe driving device 1 monitors the rotational state of the brushless DC motor 2 to get a hint of the engine 2 to detect transition to a state of synchronization loss. When the indication is detected, the drive control circuit stops 4 temporarily driving the engine 2 and brings this to a freewheeling state. Thereafter, it executes a control to drive the motor 2 to resume. Therefore, as a result, it is possible to prevent the engine from stopping, the engine from completely losing synchronization, and continuing to drive the rotation of the engine.

More specifically, the synchronization loss monitoring circuit detects 7 the speed of the motor 2 and compares the sensed speed with the normal speed of the motor 2 , When the difference between them becomes equal to or greater than a predetermined value, it detects an indication of transition to a state of synchronization loss. If it is likely that the engine 2 a synchronization loses, the speed of the motor fluctuates 2 fast. The synchronization loss monitoring circuit 7 detects a duration T60 equivalent to an electrical angle of 60 °, based on a zero-crossing point of the induced voltage of the motor 2 , Then, it compares the length of the detected duration T60 with a duration equivalent to an electrical angle of 60 ° at the normal speed. Therefore, an indication of transition to a state of synchronization loss can be easily detected.

Second embodiment

In the second embodiment of the present invention incorporated in the 4 to 6 4, a different detection method is used to predict a synchronization loss which is performed in S4 and S5. The synchronization loss monitoring circuit 7 In the second embodiment of the present invention, a logical combination therein executes based on the position signals PU ', PV', PW 'and generates a switching signal Ssw as shown in FIG 4 , (g), is shown.

The switching signal changes to a high level during a duration for which position signals PU ', PV', PW 'of any two phases are brought to a high level. It is thereby repeatedly changed to high and low levels for a duration equivalent to an electrical angle of 60 °. If the engine 2 rotating at the normal speed, repeats the output voltage of each phase of the inverter unit 3 a predetermined pattern. Therefore, the switching signal Ssw is a rectangular wave signal having a duty ratio of 50%. When a pattern different from the predetermined pattern is generated, there is a high possibility of transition to a loss of synchronization.

Therefore, the switching signal is a Ssw square wave signal with a duty cycle of 50%. When a pattern different from the predetermined pattern is generated, there is a high possibility of transition to a synchronization loss. Therefore, the synchronization loss monitoring circuit monitors 7 the output state of the switching signal to predict a loss of synchronization.

The waveform of the switching signal, which in 5 shows a case where the commutation time of the motor 2 earlier than the normal speed will be. This is an abnormal condition in which the duration of the switching signal is rapidly shortened. Conversely, as the commutation time slows down, the duration of the switching signal is increased. The waveform of the switching signal that is in 6 shows a case in which the Kommuntationszeitpunkt the engine 2 is not changed as it is expected or is regular. This is also an abnormal condition in which the duration of the switching signal is temporarily shortened. If any one of these conditions is detected for a predetermined time, which is indicative of the engine 2 is deemed to lose synchronization, a determination "YES" is made in S5.

According to the second embodiment of the present invention, the synchronization loss monitoring means performs 7 the following processing. If a duration for which a pattern of evolution of the output voltage of each phase of the inverter unit 3 does not match a predetermined pattern, becomes equal to or greater than a predetermined value, it detects an indication of transition to a state of synchronization loss. Therefore, the indication can be reliably detected.

Third embodiment

In the third embodiment of the present invention incorporated in the 7 and 8th is shown monitoring a synchronization loss monitoring circuit (synchronization loss predicting means) 7 a pattern of development of an induced voltage of each phase. For this purpose, three comparators 13 intended for each phase.

Similar to the comparators 6U . 6V . 6W In the first embodiment of the present invention comparators compare 13To . 13VM . 13WM the output voltages of the inverter unit 3 with the virtual zero potential of a reference voltage source 8M , comparators 13UH . 13VH . 13WH compare the previous output voltages with the high side threshold of a reference voltage source 8H , which is higher than the potential of the reference voltage source 8M is fixed. comparators 13UL . 13VL . 13WL Compare the previous output voltages with the low side threshold of a reference voltage source 8L which is lower than the potential of the reference voltage source 8M is fixed.

Although it is in 7 is not shown, the comparison signals PU, PV, PW, by the comparators 13To . 13VM . 13WM output as in the first embodiment of the present invention, the drive control circuit 4 fed. In 7 are the freewheeling diodes of the FETs 3a to 3f Not and is a gate drive circuit 5 ( 5a to 5f ) in a block form.

Regarding an operation common to all the phases, in the following description, numerals not "U, V or W" are attached to reference characters 7 monitors the excitation pattern for each phase based on position signals PH ', PM', PL 'obtained from comparison signals PH, PM, PL obtained from the comparators 13H . 13M . 13L be issued. At this time, it differentiates the excitation pattern into three levels: high level SH, intermediate level (high impedance) SM, and low level SL. 8th , (d), (e), (f) represent position signals PUH ', PUM', PUL 'of the U-phase.

More specifically, the high level SH and the low level SL are determined by SH = PH 'and SL = / PL'("/" represents a negation). The intermediate level SM is determined as follows. SM = (/ PH '* PM') + (/ PM '* PL')

Since the intermediate level SM can be determined, the following can be realized. When the rotation of the engine 2 is normal (steady state), the synchronization loss monitoring circuit 7 recognize that the excitation pattern in each of the U, V and W phases transition from a state 1 to a state 6, tabulated below, every 60 ° of an electrical angle. Here, "M" represents a high impedance. Status 1 2 3 4 5 6 U L L M H H M V H M L L M H W M H H M L L

Consequently, the synchronization loss monitoring circuit monitors 7 whether the previous cycle is repeated from state 1 to state 6 in a proper (normal) pattern. If there is a deviation from the correct pattern, it is determined that there is an indication of synchronization loss.

According to the third embodiment of the present invention, the synchronization loss monitoring circuit performs 7 the following processing.

Upon detecting a pattern of developing an output voltage of each phase for the motor 2 The output voltage is differentiated into three levels: high level, low level and non-energized level (intermediate level). Therefore, the state of rotation of the motor can be clearly monitored.

In the third embodiment of the present invention, the intermediate level SM can be further divided into two levels to divide a pattern of an output voltage. SMH = / PH '· PM' SML = / PM '· PL'

Furthermore, the comparators 13To . 13VM . 13WM can be eliminated, and a determination can be made by SM = / PH '· PL'

Regarding the states 1 to 6 may be the following measure carried out become. Only a duration "M", for which a high impedance is achieved, is selectively selected and it is being monitored whether the development pattern for go through the duration "M" normally will or not.

Fourth embodiment

In the fourth embodiment of the present invention incorporated in the 9 and 10 is a lathe driving device 1 constructed such that a synchronization Ver Loss control circuit or synchronization loss predictor 7 detects a current that is a motor 2 is supplied, and predicts a synchronization loss based on the state of a fluctuation of the current. The drain and the gate of the FET 3a the inverter unit 3 are connected to the drain and gate, respectively, of a current sense N-channel power MOSFET 3 connected. They are switched on / off together by a common gate signal. The FETs 3a . 23 are set so that when they are turned on, the ratio of currents respectively flowing therethrough is 100: 1 to 5000: 1 and so on.

The source of the FET 23 is over a diode 24 , the collector / emitter of an NPN transistor 25 and a resistance 26 connected to a ground wire and is further connected to a terminal (+) of an operational amplifier 27 connected. The connection (-) of the operational amplifier 27 is with the source of the FET 3a and its output terminal is connected to the base of the transistor 25 connected. The emitter of the transistor 25 is the terminal (+) of a comparator 28 connected and the connection (-) of the comparator 28 is for supplying a reference voltage for comparison with a voltage source 29 connected. The comparator 28 is configured such that its output signal into the drive control circuit 4 is entered.

If the FETs 3a . 23 are switched on simultaneously, drain currents corresponding to their current ratio, each passed through this. In this case, the source voltages become due to the operation (imaginary short circuit) of the operational amplifier 27 equal to each other. Even if the resistance 26 with the current detection FET 23 Therefore, their current ratio is kept as determined.

The current supplied when the FET 23 is turned on, flows by means of the diode 24 and the transistor 25 to the resistance 26 , The terminal voltage of the resistor 26 is through the comparator 28 with the reference voltage of a voltage source 29 compared. As the level of the former becomes higher, the comparator changes 28 the output signal to a high level.

As it is in 10 As shown, a current fluctuates with the motor 2 is supplied before it goes to a state of a synchronization loss. However, the current shown in this figure is not current flowing from the FET 23 but the whole through a DC power supply line. If the engine 2 in a normal (steady state), the current hardly fluctuates and is substantially constant. If any error in the rotation of the motor 2 and its output torque fluctuates or fluctuates due to any similar circumstances, a fluctuation of the current is also increased in conjunction therewith. Therefore, the synchronization loss monitoring circuit says 7 a synchronization loss by detecting a current fluctuation (increase) at this time beforehand.

More specifically, if the comparator 28 the output to the high level changes, the drive control circuit 4 triggered by this level change and brings the engine 2 to a freewheeling state. This processing corresponds to the processing of S4 to S6 in FIG 2 ,

According to the fourth embodiment of the present invention, the synchronization loss monitoring circuit detects 7 a current belonging to the engine 2 is supplied. When a fluctuation of the current becomes equal to or greater than a predetermined value, it detects an indication of a loss of synchronization. Therefore, a loss of synchronization can be reliably predicted.

In the first to fourth embodiments The present invention can provide a loss of synchronization by Combining synchronization loss prediction methods in various embodiments and applying an OR link be predicted. The embodiments are not only to those for driving the drive motor of an electric vehicle applicable. You can are widely used in applications in which it is difficult is a brushless one DC motor to stop by a loss of synchronization, when the motor is driven by a sensorless process.

Fifth embodiment

In the fifth embodiment of the present invention, which is shown in FIG 11 is a lathe driving device 1 designed to drive a motor for driving, for example, a mini disk or MD or a hard disk drive or HDD. The lathe driving device tung 1 is similar to that of the first embodiment of the present invention ( 1 ), but has no synchronization loss monitoring circuit. However, a drive control circuit includes 4 an internal counter (not shown) for measuring an interval between the edges of the position signals PU ', PV', PW 'to measure a zero-crossing period T60 corresponding to an electrical angle of 60 °, as shown in FIG 12 , (g), is shown. The zero crossing time point of the induced voltage has a phase delay of an electrical angle of 30 ° from an appropriate energization time. Therefore, the drive control circuit generates 4 a commutation pattern by compensating the phase delay. The time interval corresponding to the electrical angle of 30 ° can be determined as T60 / 2. The waveforms of the output voltage of the inverter unit 3 that are observed when the engine 2 turns, are in 12 shown. When the three-phase motor 2 its voltage waveforms are similar to those of the first embodiment of the present invention ( 3 ).

Gate drive circuits 5a . 5c . 5f a high side are similar to each other. For example, the gate drive circuit is 5a built as it is in 13 is shown. The drain and the gate of the FET 3a that is the inverter unit 3 are each connected to the drain and the gate of a current sense N-channel power MOSFET 107 connected. They are switched on and off simultaneously by a common gate signal. The FETs 3a . 107 are set so that when they are turned on, the ratio of currents flowing therethrough is, for example, 100: 1 to 5000: 1 and so on.

Between the power supply line (VB) and a ground line is a series of resistors 111 and 112 and an N-channel MOSFET 113 connected. The common connection point of the resistors 111 and 112 is at the base of a PNP transistor 114 connected and the emitter of the transistor 114 is connected to the power supply line. The collector of the transistor 114 is over a boost circuit area 115 and a diode 116 with the gates FETs 107 and 3a connected.

Between the anode of the diode 116 and the source of the FET 107 are two zener diodes 117 and 118 connected in series such that they are in an opposite direction to each other. The amplification circuit area 115 is used to perform a voltage boosting operation to achieve a gate voltage required to drive the N-MOSFETs 3a and 107 to head for a high page. It is constructed of a conventional charge pump circuit made up of a combination of diodes and capacitors.

The source of the FET 107 is over a diode 119 , the collector / emitter of an NPN transistor 120 and a resistance 121 connected to the ground wire and is still connected to the terminal (+) of an operational amplifier 22 connected. The connection (-) of the operational amplifier 122 is with the source of the FET 3a and its output terminal is connected to the base of the transistor 120 connected. The emitter of the transistor 120 is connected to the terminal (-) of a comparator 123 connected and the terminal (+) of the comparator 123 is for supplying a reference voltage for comparison with a voltage source 124 connected. The output signal of the comparator 123 is about a filter 125 any input terminal of an AND gate 126 fed.

The AND gate 126 is constructed such that the following is realized. When the output signal of the comparator 23 At a high level, there is a gate signal UH supplied from the drive control circuit 4 is output to the FETs 3a and 107 out. When the output signal transitions to a low level, it stops the output of the gate signal UH. When a gate drive signal of a high level is applied to the FET 113 is fed, the FET 113 is turned on and as a result the transistor 114 also turned on. Therefore, a gate drive voltage with respect to the source of the FET 107 to the gates of the FETs 3a and 107 created, with the FETs 3a and 107 be turned on again.

The FETs 3a and 107 are in a current mirror configuration and their source voltages are due to the operation (imaginary short circuit) of the operational amplifier 122 equal to each other. Even if the resistance 121 with the current detection FET 107 Therefore, their current ratio is kept as determined. If the FETs 3a and 107 be turned on, the current flowing to the FET 107 is supplied by means of the diode 119 and the transistor 120 to the resistance 121 , The terminal voltage level of the resistor 121 is through the comparator 123 with the reference voltage of the voltage source 124 compared.

The time constant of the filter 125 is set so that the following will be realized as later in 14 is shown. Noise at a frequency that is higher than the off time Toff when on control current from the AND gate 126 is interrupted is cut off.

An operation of this embodiment of the present invention will be described with reference to FIG 14 to 17 described. The drive control circuit 4 performs processing that is in 17 which is similar to that of the first embodiment of the present invention ( 2 ) when the engine 2 is started. Part of the processing involves being executed by hardware. The drive control circuit 4 excites the turns 2U . 2V . 2W of the motor 2 through a DC current to position its rotor (S1). Then the turns become 2U . 2V . 2W is energized according to a predetermined Kommutationsmuster and therefore the engine 2 started by a forced commutation (S2).

The drive control circuit 4 is constructed such that, when forced commutation is performed, it performs an increased excitation, that is, excitation at an increased or anticipatory angle. That is, with respect to the position of the rotor positioned at S1, a commutation is made at a time leading by 30 ° from a normal appropriate commutation time. By carrying out the leading excitation, when an engine 2 is started, the torque of the engine 2 reduced. Therefore, the effect of suppressing the occurrence of excessive speed is achieved.

When the number of revolutions of the engine 2 Increased to some extent, after it has been started, may induce an induced voltage in the windings 2U . 2V . 2W is being developed. As a result, the drive control circuit changes 4 the driving method for the motor 2 to a sensorless mode (S3). That is, a commutation pattern signal for the inverter unit 3 is generated on the basis of the comparison signals PU, PV, PW, and gate signals UH to WL become the gates of the individual FETs 3a to 3f output. The excitation phase angle in the sensorless mode advances 30 ° with respect to the zero crossing point of an induced voltage.

If a forced commutation is performed in S2, the output of the comparator changes 123 in the gate drive circuit 5a in accordance with the current supplied by the FET 107 captured the engine 2 is supplied. The reference voltage of the voltage source 124 is set to such a level that an excessive current that is supplied when the engine 2 is started, as it is limited in 14 is shown. In this embodiment, it is set to a voltage equivalent to, for example, a current of 2A.

15 represents voltages Vu to Vw of the respective phases and a motor current Im which is observed when a current supplied when the motor 2 actually started, is limited to 2 amps. As it is in 15 , (d), is the current Im, which is conducted when the motor 2 normal turns, less than 1 A. The limit level of 2 A provides an upper limit, which is fixed with respect to a current at a too high level, which is conducted only at the time of starting.

When the motor current Im exceeds the limit level, the output current of the comparator becomes 123 changed to a low level and prevents the AND gate 126 the output of the gate signal UH. Therefore, the FETs 3a . 3c . 3e a high side of the inverter unit 3 turned off and will be an excitement of the engine 2 stopped. As a result, a detected current value decreases and the output current of the comparator returns 123 back to a high level. In conjunction with this takes the AND gate 126 the output of the gate signal UH back on and becomes the motor 2 excited.

A change in the detected current will occur at a predetermined gradient through the inductance of the turns 2U to 2W of the motor 2 intended. As it is in 14 is shown, the detected current changes at the time of start similar to a sawtooth wave near the limit level, and the current is restricted.

16 represents a change of the start time (ms) observed in a case where the current is not restricted when the motor 2 is started, and in cases where a current is limited to the limit levels of 1 A, 2 A and 4 A. The "start time" mentioned herein is defined as a time until the rotation of the motor reaches 90% of its normal number of revolutions (for example, 10000 rpm). The horizontal axis represents the initial position (degree) of the runner.

If this starting position changes, the start time may change as well. If the limit level is set to 1 A, this limit level is too low and a required starting torque can not be obtained. The start time is 100 ms over the entire range of an initial position and this is much bigger as, for example, 70 ms, which is a standard value required by articles. In the case of "no limit" and in cases where the current is limited to 2 A and 4 A, this start time is significantly lower than the previously required standard over the entire range of this initial position.

The three cases, in which the start time is shorter as the required standard is evaluated. In that case At the limit level of 2 A, the start time is 25 ms over the entire range of the initial position. In the cases of "no limit" and the limit level 4 A can make the start time shorter than in the case of the limit level of 2 A depending on the initial position be. However, the worst values (30 ms) are greater in both cases in the case of the limit level of 2 A. Therefore, the case of the limit level of 2 A as preferred for Products.

In In some other examples of a measurement, the start time is in the Case "no limit" in which a stream strongly fluctuates, and in the case of the limit level of 4 A longer than in the case of the limit level of FIG. 2A.

According to this embodiment of the present invention, the following is realized when the lathe driving device 1 the brushless DC motor 2 by a forced commutation starts. The gate drive circuits 5 limit the flow of the turns 2U to 2W of the motor 2 is supplied to an upper limit level set higher than a level to which a current is supplied when the engine 2 is in a normal or stable state of rotation. Therefore, it is possible to suppress the excessive speed to shorten a start time without preventing the stable rotation of the motor.

When the drive control circuit 4 Performs a compulsory commutation after a positioning of the rotor, it carries out a control such that the excitation phase angle for the windings 2U to 2W advanced by a predetermined amount. Therefore, a start time can be further shortened.

The fifth embodiment The present invention can be configured in many ways. For example, you can the limit levels for start currents and the normal number of revolutions expediently according to the nominal value of an engine, which is used or the like can be configured. A premature excitement in a forced commutation can be performed as required. The lathe driving device is not only to those to drive the motor of a minidisk or MD or a hard disk drive or HDD applicable. It can be widely used in applications in which it is difficult, the rotation of the motor through Synchronization loss stop when the brushless DC motor is controlled by the sensorless process.

Sixth embodiment

In the sixth embodiment according to the present invention disclosed in 18 is a lathe driving device for a brushless DC motor 2 provided, which may be provided in a vehicle as a lathe.

The motor 2 is a three-phase motor and an actuator for a fuel pump for an internal combustion engine, which is installed in a motorcycle. The three phases (U-phase, V-phase, W-phase) of the brushless motor 2 are with an inverter 12 connected. The inverter 12 is a three-phase inverter and sets the voltage of a battery 214 to the three phases of the brushless motor 2 at. To connect a line between each of the three phases and the positive pole and negative pole of the battery 214 is the inverter 12 such that it includes a parallel circuit unit comprising: switching elements SW1, SW2 (U-phase leg), switching elements SW3, SW4 (V-phase leg), and switching elements SW5, SW6 (W-phase leg). The connection point between the switching element SW1 and the switching element SW2 connected in series is with the U-phase of the brushless motor 2 connected. The connection point between the switching element SW3 and the switching element SW4 connected in series is with the V-phase of the brushless motor 2 connected. The connection point between the switching element SW5 and the switching element SW6 connected in series is with the W phase of the brushless motor 2 connected. These switching elements SW1 to SW6 are each connected in parallel with free-wheeling diodes D1 to D6.

The Switching element SW1, SW3, SW5 is a high side of each branch composed of a P-channel MOS transistor and the switching element SW2, SW4, SW6 of a low side of each branch is one N-channel MOS transistor constructed. The freewheeling diodes D1 to D6 are from parasitic diodes the MOS transistors built up.

A drive control circuit 220 operates the inverter 12 and thereby controls the output of the brushless motor 2 , More specifically, the drive control circuit includes 220 a driver 222 , a voltage detecting device 228 and a shift control device 227 , The voltage detection device 228 detects the voltage VB of the battery 214 ,

The switching control device 227 switches the switching elements SW1 to SW6 through the driver 222 in and out. In this example, it mainly performs switching control by a 120 ° excitation method. More specifically, the switch controller detects 227 the following using the instant at which the terminal voltages Vu, Vv, Vw of the respective phases of the brushless motor 2 equal to the induced voltage. It detects a time (zero-crossing time) at which the induced voltage equals the virtual zero point voltage (reference voltage Vref) of the brushless motor 2 becomes. Then, it changes the operation of the switching elements SW1 to SW6 at the timing (specific time) delayed from the zero-crossing time by a predetermined electrical angle, for example, 30 degrees (°). However, when a current (amount of arousal), which is the brushless motor 2 is limited, the duration for which the switching elements SW1 to SW6 are turned on, not set to a 120 ° duration. Instead, PWM control is performed during this duration.

The switching control device 227 may be constructed of a logic circuit or may be constructed of a central processing unit and a memory unit for storing a program.

2 FIG. 12 illustrates the manner in which a shift control under normal conditions is executed. More specifically, (a) represents the transition of the terminal voltages Vu, Vv, Vw of the motor 2 group; represents (b) the transition of comparison signals Uc, Vc, Wc; represents (c) the transition of zero-crossing detection signals; represents (d) the transition of the values of different counters; and (e) represents the transition of enable signals for the switching elements SW1 to SW6. The enable signals shown in (e) include enable signals U +, V +, W + for the high side switching elements SW1, SW3, SW5 of the respective branches and Activation signals U-, V-, W- for the switching elements SW2, SW4, SW6 a low side. In this example, the switching elements SW1, SW3, SW5 of a high side of the respective branches are P-channel transistors; therefore, the durations for which these activation signals U +, V +, W + are at logic L are the durations for which they are turned on.

Upward solid Lines in (d) show the value Cm of the zero-crossing measurement counter for measurement a time interval between adjacent zero crossing times at. As shown in the figure, the counter becomes initialized at each time the zero crossing time occurs, and starts a time counting process New. A time interval between adjacent or subsequent ones Zero crossing times has a correlation to a speed. Because of this, the value of the counter looks just before it initializes (the maximum value of the counter) a parameter, which has a correlation to the speed.

Down solid Lines in (d) show the value Cs of a counter for specifying a particular counter Time that counts a time which is required until a zero crossing time equals a certain one Time, thereby defining a particular point in time. The counter to set a certain time takes the value of the counter an initialization as its initial value at the zero-crossing time on and decrements it. Then he sets the time to which the value zero is fixed as a certain time. To this Time, the following process is performed.

If the time interval between the zero crossing time and the determined one Time is 30 ° is for example, the decrement rate is twofold the incrementing speed of the measuring counter. Considering its that the time interval between adjacent zero crossing times 60 °, it may be considered that it allows this setting, the time at which the Cs value of the counter sets a particular Time zero becomes 30 ° from the Delay zero crossing time.

The two-dot chain lines in (d) indicate the value Cp of a release start counter. The release start counter determines the time of commencement of a duration (enabled duration) for which detection of the zero-crossing time is enabled based on the comparison of the terminal voltages Vu, Vv, Vw and the reference voltage Vref. The approved duration is provided to avoid the following event and for other similar purposes. In a period for which a current is supplied through the diodes D1 to D6, the terminal voltages Vu, Vv, Vw become equal to the reference voltage Vref, and this circumstance is erroneously determined as the zero-crossing time. This counter also takes the value of the counter before initialization as its initial value at the zero-crossing time and decrements it. Then he sets the time at which the value is zero, as the time of the beginning of a shared period. For example, when the time of the start of a released period is set to be 45 ° from the zero-crossing time, the decrementing speed may be set to 3/2 times the incrementing speed of the measuring counter.

The single-dashed lines in (d) show the value Cps of a counter to set a shared duration to determine the previous one shared duration. If the value of the enable start counter is zero is, takes the counter To set a shared duration, preset the value of the meter the previous initialization as its initial value and decrements it. It sets the duration until the value is zero, as the released duration. If the released duration is a duration of 30 °, For example, the decrement rate can be twofold be set the incrementing speed of the measuring counter.

In the duration, for which is the value of the meter to set a shared duration is not less than zero, the comparison signals Uc, Vc, Wc are made valid. When the comparison signals Uc, Vc, Wc are inverted in this period, the zero-crossing detection signals the corresponding phases inverted. At the zero crossing time, to which the zero-crossing detection signal is inverted becomes the decrementing of the counter to set a specific time. If its value Is zero, the operation of the switching elements SW1 to SW6 is changed.

The switching control processing in this embodiment will next be described with reference to FIG 20 and 21 described. The processing of 21 for setting the counter values with respect to the previous four counters is repeated by the switching control circuit 220 for example, executed in a predetermined cycle.

The Series of processing is carried out as follows. In S10 it is checked if the Value Cps of the counter to set a shared duration equal to or greater than Zero or not. If it is determined that the value is not is less than zero, it is checked in S12 if any of the comparison signals Uc, Vc, Wc changed or has been inverted or not. This processing serves for detecting the zero-crossing time. If it is determined in S12, that any of the comparison signals has been changed, the current value Cm of the measuring counter as the maximum value of the counter taken in S14. In S16 below, the maximum value of the counter is called the values Cs and Cps of the counter to set a specific time and the counter to Set a shared duration. In S18 the measuring counter is initialized (Cm = 0).

If in S10 or S12 a negative determination is made, becomes the meter (Cm) increments in S20. At the same time the release start counter (Cp), the counter (Cps) to set a shared duration and the counters (cs) decremented to set a particular time. In S22 it will be checked below if the Value Cp of the release start counter Zero or not. This processing serves to check that whether it is at the time of commencement of a shared period or not. If it is determined that the value Cp of the enable start count is zero is the maximum value Cm of the counter as the value Cps of the counter to set a released duration in S24. Here in further course is detecting the zero crossing time on the Basis of the comparison signals Uc, Vc, Wc released until the Value Cps of the counter to set a shared duration is zero.

If in S22 a negative determination is made or if the processing finished by S18 or S24, this series becomes a processing finished again.

Changing the state of the switching elements SW1 to SW6 to ON is carried out by the processing described in FIG 21 is shown. This processing is performed by the drive control circuit 220 for example, repeatedly executed in a predetermined cycle.

In S30, it is checked whether the value Cs of the counter for setting a specific time is zero or not. This processing is for determining whether the specific time has occurred or not. When it is determined that the value Cs of the counter for setting a specific time has become zero, the operating state of the switching elements SW1 to SW6 is changed in S32. In this example, the operating state changes as follows. When the switching elements SW1, SW4 are ON before being changed, the operating state is changed so that the switching elements SW1, SW6 are turned ON. When the switching elements SW1, SW6 are ON before being changed, the operating state is changed so that the switching elements SW3, SW6 are turned ON. If the switching elements SW3, SW6 before a Än If ON, the operating state is changed so that the switching elements SW2, SW3 are turned ON. When the switching elements SW2, SW3 are ON before being changed, the operating state is changed so that the switching elements SW2, SW5 are turned ON. When the switching elements SW2, SW5 are ON before being changed, the operating state is changed so that the switching elements SW4, SW5 are turned ON. When the switching elements SW4, SW5 are ON before being changed, the operating state is changed so that the switching elements SW1, SW4 are turned ON.

In this embodiment The present invention is a switching control by a 120 ° excitation method by bringing the timing to which the switching elements SW1 changed to SW6 become a one-to-one correspondence executed at a zero crossing time.

In a vehicle is the voltage of the battery 214 prone to waver. When the voltage of the battery 214 varies, the speed of the brushless motor changes 2 , At this time, the time interval between adjacent zero-crossing times (the maximum value of the counter) may not accurately represent the speed to one of the current zero-crossing time. Therefore, a released period can not be set in a desired electrical angle range depending on the previous time interval. Because of this, there is, if the brushless motor 2 for example, a possibility that the zero-crossing time occurs before the enabled duration. If the brushless motor 2 For example, there is a possibility that the zero-crossing time will occur after the enabled period.

In this embodiment of the present invention, to cope with it, the setting of a permitted duration corresponding to a change in the rotational speed of the brushless motor 2 corrected. Taking into account that a speed by changing the voltage VB of the battery 214 In particular, the setting of a permitted duration corresponding to a change in the voltage of the battery is changed 214 corrected.

This fix for setting a shared duration is performed as described in 22 is shown. This processing is done by the switching control circuit 220 for example, repeatedly executed in a predetermined cycle.

In S40, the voltage VB of the battery becomes 214 detected. In S42, it is subsequently checked whether a change ΔVB of the voltage VB of the battery 214 is equal to or greater than a first threshold α or not. This processing is for determining whether or not the present situation is a possible situation in which the following occurs. As a result of that the brushless motor 2 is accelerated, the zero crossing time occurs before the enabled duration, which is determined by the speed. This threshold α is set on the basis of the minimum value at which the previous situation occurs. When it is determined that the change .DELTA.VB is equal to or higher than the threshold value .alpha., The count speed of the release start counter and the permissive duration set counter is increased in S44. As it is in 23 In particular, the count speed is increased with an increase in the change ΔVB. In this example, the following action can be taken. A correction coefficient is predetermined for each discrete value of a change ΔVB, and the count speed is changed stepwise. Instead, the count speed may be continuously changed in accordance with the change ΔVB. This processing serves to realize the following in a situation where the rotational speed of the brushless motor 2 is increased. A released duration is set substantially in the same electrical angle range as when the speed is constant.

If the change ΔVB is smaller than the threshold α, it is determined in S46 whether or not the change ΔVB is equal to or lower than a second threshold β. This processing serves to determine if the current situation is a possible situation in which the following occurs. As a result of that the brushless motor 2 is delayed, the zero-crossing time occurs after a released period, which is determined by the speed. This threshold β is set on the basis of the maximum value at which the previous situation occurs. When it is determined that the change ΔVB is equal to or smaller than the threshold value β, the count speed of the release start counter and the permissive duration set counter is decreased in S48. As it is in 23 In particular, the counting speed is reduced with a decrease in the change ΔVB. The counting speed is reduced when the change takes a negative value and with an increase in its maximum value. This processing serves to realize the following in a situation where the rotational speed of the brushless motor 2 is reduced. A released duration is set substantially in the same electrical angle range as when the speed is constant.

If it is determined in S46 that the change ΔVB is larger than the threshold value β, or if the processing of S44 or S48 is finished, this series of processing is terminated once. According to the above processing, a zero-crossing time can be appropriately detected even when the rotational speed of the brushless motor 2 by changing the voltage VB of the battery 214 will be changed.

There are cases where the speed of the brushless motor 2 becomes excessively high or low on the basis of a detected value of zero-crossing timing. This leads to a state in which it is difficult to suitably the rotational state of the brushless motor 2 to control (state of loss of synchronization). In this condition, it is desirable to use the brushless motor 2 to restart. Before this condition occurs, the speed of the brushless motor becomes 2 continuously increased or decreased. For this reason, a phenomenon that the zero-crossing time deviates from a permitted duration tends to occur more than once in succession. In this embodiment of the present invention, the following processing is subsequently performed when the number of times when the zero-crossing time either continuously occurs before or after a released period is equal to or higher than a threshold value. The state of rotation of the brushless motor 2 is determined to be abnormal and processing is performed to the brushless motor 2 to restart.

Here in the course of this processing is with reference to 24 and 25 described. 24 represents the processing for coping with an anomaly that the zero-crossing time occurs before the released period. This processing is performed by the drive control circuit 220 for example, repeatedly executed in a predetermined cycle.

In This processing is checked in S50 whether the Value Cp of the release start counter Zero or not. This processing is to determine whether it is the time of the beginning of the approved duration or not. If it is determined that the value Cp of the release start counter is zero is, it is determined in S52 whether the logic value of the corresponding Comparison signal Uc, Vc, Wc is normal or not. This processing serves for determining if the zero crossing time is before the enabled duration occurred or not.

As it is in 19 is shown, the zero crossing time occurs every 60 °. Whether the induced voltage of each phase crosses the reference voltage Vref in its rise or fall operation has a 360 ° periodicity. For this reason, in which phase the current zero-crossing time occurs and whether the induced voltage of that phase is in its ramp-up or fall operation can be identified according to: in which phase the previous zero-crossing time has occurred and if the induced voltage of that phase is in its ramp-up process or waste process. More specifically, in the example that is in 19 Assuming that the zero-crossing time at which the terminal voltage Vv of the V-phase crosses the reference voltage Vref in its decay operation is assumed. In this case, it can be predicted that the next zero-crossing time will occur in the U-phase and the zero-crossing time will occur in the rising operation of the terminal voltage Vu of that phase. For this reason, it can be considered that in the enabled period, the U-phase comparison signal Uc is inverted from logic L to logic H. Meanwhile, when the U-phase comparison signal Uc has already been at logic H, when the value of the enable start counter is zero, it may be determined that the zero-crossing time has occurred before the enabled period.

If the logic value of a comparison signal of a phase in which it It is assumed that the zero crossing time in the shared Duration occurs when it is determined to be abnormal, processing proceeds S54 on. In S54, it is determined that the zero-crossing timing is excessively advanced. Then the value of the counter for the Values Cp Re of the release start counter and the counter Replaced Cs to set a specific time. Therefore, will the time at which the value Cp of the release start counter is zero, taken as a zero crossing time at a certain time and the like. This processing is to avoid the following even if an erroneous determination is made in S52: the setting of a certain time and the like is excessive of a normal time shifted. In S56 below is the measuring counter initialized.

In S58, it is checked whether the determination of excessive overrun has continued or not, that is, the previous zero-crossing time has also occurred before the enabled period. If the previous zero-crossing time also occurred before the enabled period, an excess overflow counter is incremented in S60. The excessive overflow counter is arranged to count a number of times Ca at which the zero-crossing time occurs continuously before the enabled duration. In S62, it is subsequently checked whether or not the value Ca of the excess overflow counter has become equal to a threshold value MAX. This processing is for bestim whether the rotational state of the brushless motor is abnormal or not and it is difficult to control the rotational state. The threshold value MAX is set to a value greater than an estimated number of times, at which the zero-crossing time before the enabled duration at a start of the brushless motor 2 will occur. When it is determined that the value of the excess overflow counter is equal to the threshold MAX, restart processing (recovery processing) for the brushless motor becomes 2 executed in S64. At the same time, the excess overflow counter is initialized.

If a negative determination is made in S58, the excess overflow counter is initialized in S66. This processing is to avoid the following when the zero-crossing time due to a start of the brushless motor 2 , the influence of noise or the like, a load change or the like occurs before the released period. The value Ca of the excess overflow counter is counted and finally determined to be abnormal, and restart processing is executed. If a negative determination is made in S50 or S52, or if the processing of S64 or S66 is finished, this series of processing is terminated once.

25 represents the processing for coping with an anomaly that the zero-crossing time occurs after a released period. This means that the zero crossing time is delayed. This processing is repeated by the switching control circuit 220 for example, repeatedly executed in a predetermined cycle. This series of processing is performed as follows.

In S70, it is checked whether the value Cps of the counter for setting a permitted duration is zero or not. This processing is for determining whether or not it is the time of the end of the enabled period. If it is determined that the value Cps of the permitted period setting counter is zero, it is determined in S72 whether in S52 in FIG 24 an affirmative determination has been made or not and the zero-crossing time has already been detected. This processing is for determining whether the zero-crossing time occurs after the enabled period or not. More specifically, if the zero-crossing time has not occurred before the released period nor occurs in the enabled period, it may be expected to be delayed and to occur after the enabled period.

When a negative determination is made in S72, processing similar to that of S54 to S66 in FIG 24 in S74 to S86 in which a delay is counted as the delay value Cd using a delay counter.

In this embodiment of the present invention, a synchronization loss state is determined on the basis that the zero-crossing time continuously occurs before or after the released period. For this reason, a state in which the zero-crossing time is accidentally caused by the influence of noise or the like to deviate from the enabled duration can be discriminated from a synchronization loss state. Furthermore, setting is performed such that the enabled duration is disregarded from a fluctuation of the voltage of the battery 214 at a predetermined electrical angle. Therefore, a synchronization loss state can be prevented from being caused due to a fluctuation of the voltage of the battery 214 is determined.

• (1) Wenn die Anzahl von Zeiten, zu denen die Nulldurchgangszeit kontinuierlich entweder vor oder nach der freigegebenen Dauer auftritt, gleich oder höher als ein Schwellwert MAX wird, wird der bürstenlose Motor 2 als anormal bestimmt. Daher kann, wenn eine Drehschwankung vorübergehend aufgrund einer Laständerung oder dergleichen oder irgendein anderes ähnliches Ereignis auftritt, verhindert werden, dass dies als anormal bestimmt wird.
• (2) Wenn es bestimmt wird, dass die Nulldurchgangszeit vor der freigegebenen Dauer aufgetreten ist, wird angenommen, dass die Zeit des Beginns der freigegebenen Dauer die Nulldurchgangszeit ist, um einen bestimmten Zeitpunkt festzulegen. Daher kann es auch dann, wenn eine Bestimmung fehlerhaft ist, vermieden werden, dass ein bestimmter Zeitpunkt zu unzweckmäßig festgelegt wird.
• (3) Wenn es bestimmt wird, dass die Nulldurchgangszeit nach der freigegebenen Dauer auftritt, wird angenommen, dass die Zeit des Endes der freigegebenen Dauer die Nulldurchgangszeit ist, um den bestimmten Zeitpunkt festzulegen. Daher kann es auch dann, wenn eine Bestimmung fehlerhaft ist, vermieden werden, dass ein bestimmter Zeitpunkt zu unzweckmäßig festgelegt wird.
• (4) Auf der Grundlage des Werts eines Vergleichssignals zu der Zeit des Beginns der freigegebenen Dauer wird es überprüft, ob die Nulldurchgangszeit vor der freigegebenen Dauer aufgetreten ist oder nicht. Auf der Grundlage des Vorhandenseins oder Nichtvorhandenseins der Änderung eines Vergleichssignals in der freigegebenen Dauer wird es überprüft, ob die Nulldurchgangszeit nach der freigegebenen Dauer auftritt oder nicht. Daher können diese Bestimmungen zweckmäßig durchgeführt werden.
• (5) Gemäß einer Änderung der Drehzahl des bürstenlosen Motors 2 wird das Festlegen der freigegebenen Dauer auf der Grundlage der Werte bezüglich des Freigabestartzählers und des Zählers zum Festlegen einer freigegebenen Dauer korrigiert. Dies ermöglicht es, genau eine freigegebene Dauer in einem erwünschten elektrischen Winkelbereich unberücksichtigt irgendeiner Änderung der Drehzahl festzulegen und die Robustheit eines Steuerns bezüglich des Drehzustands des bürstenlosen Motors 2 zu verbessern.
• (6) Eine Änderung einer Drehzahl wird durch eine Spannungsänderung ΔVB der Spannung VB der Batterie 214 erfasst. Daher kann eine Änderung einer Drehzahl zweckmäßig erfasst werden. Genauer gesagt wird der folgende Vorteil durch Erfassen einer Drehzahl auf der Grundlage der Spannung VB der Batterie 214 vorgesehen. Wenn der Drehzustand unberücksichtigt einer Schwankung der Spannung VB der Batterie 214 anormal wird, wird das Festlegen der freigegebenen Dauer unverändert gehalten. Aus diesem Grund kann das Festlegen der freigegebenen Dauer auf der Grundlage des Zeitintervalls zwischen den Nulldurchgangszeiten lediglich korrigiert werden, wenn die Spannung VB der Batterie 214 schwankt.

According to the sixth embodiment of the present invention, the following advantages can be provided.

• (1) When the number of times when the zero-crossing time continuously occurs either before or after the released period becomes equal to or higher than a threshold value MAX becomes the brushless motor 2 determined as abnormal. Therefore, when a rotation fluctuation occurs temporarily due to a load change or the like or any other similar event, it can be prevented from being determined as abnormal.
• (2) When it is determined that the zero-crossing time has occurred before the enabled period, it is assumed that the time of the beginning of the enabled period is the zero-crossing time to set a certain time. Therefore, even if a determination is erroneous, it can be avoided that a certain time is set improperly.
• (3) When it is determined that the zero-crossing time occurs after the released period, it is assumed that the time of the end of the enabled period is the zero-crossing time to set the specific time. Therefore, even if a determination is erroneous, it can be avoided that a certain time is set improperly.
• (4) Based on the value of a comparison signal at the time of the start of the enabled period, it is checked whether or not the zero-crossing time has occurred before the enabled period. On the basis of the presence or absence of the change of a comparison signal in the Shared duration, it is checked whether the zero-crossing time occurs after the shared duration or not. Therefore, these provisions can be carried out expediently.
• (5) According to a change in the speed of the brushless motor 2 the setting of the enabled duration is corrected on the basis of the enable start counter and the release period setting counter. This makes it possible to set precisely one released period in a desired electrical angle range regardless of any change in the rotational speed and the robustness of controlling the rotational state of the brushless motor 2 to improve.
• (6) A change of a rotational speed is caused by a voltage change ΔVB of the voltage VB of the battery 214 detected. Therefore, a change in a rotational speed can be suitably detected. More specifically, the following advantage is obtained by detecting a rotational speed based on the voltage VB of the battery 214 intended. When the rotational state is disregarded from a fluctuation of the voltage VB of the battery 214 becomes abnormal, setting the shared duration is kept unchanged. For this reason, setting the enabled period based on the time interval between the zero-crossing times can be corrected only when the voltage VB of the battery 214 fluctuates.

Seventh embodiment

In the seventh embodiment of the present invention, the acceleration of the brushless motor 2 is detected on the basis of a detected value of a zero-crossing time, and the setting of a released period is corrected on the basis of this acceleration. 26A represents a processing for correcting the setting of a permitted duration. This processing is performed by the drive control circuit 220 for example, repeatedly executed in a predetermined cycle. The series of processing is performed as follows.

In S90, the acceleration A of the brushless motor becomes 2 calculated on the basis of the previous maximum value of the counter and the present maximum value of the counter. The maximum value can be determined based on the Cm ( 19 ) of the measuring counter. The maximum value of the counter is used to count the time interval between zero-crossing times and has a correlation to the instantaneous speed between zero-crossing times. For this reason, these two values, the previous maximum value of the counter and the present maximum value of the counter, are values at two different times with respect to an instantaneous speed. For this reason, the acceleration A can be calculated therefrom.

In S92, it is subsequently checked whether the acceleration A is equal to or higher than a threshold Amax or not. This processing is to determine whether or not the zero-crossing time occurs before the enabled duration determined by its rotational speed as a result of the brushless motor 2 is accelerated. When it is determined that the acceleration is equal to or higher than the threshold Amax, the same processing as that of S44 in FIG 22 executed in S94. That is, the count speed of the enable start counter and the set period setting counter is increased with increasing the acceleration A as shown in FIG 26B is shown.

If a negative determination is made in S92, it is determined in S96 whether or not the acceleration A is equal to or smaller than a threshold value Amin. This processing is to determine whether or not the zero-crossing time occurs after the enabled duration, which is determined as a result of its rotational speed, that of the brushless motor 2 is delayed. When it is determined that the acceleration is equal to or smaller than the threshold value Amin, the same processing as that of S48 in FIG 22 executed in S98. That is, the count speed of the release start counter and the permitted period setting counter is decreased with decreasing the acceleration A. The counting speed is reduced when the acceleration takes a negative value and with increasing its absolute value.

If in S96 a negative determination is made or if the processing finished by S94 or S98, this series is a processing ended once.

• (7) Auf der Grundlage des Ergebnisses eines Erfassens der Nulldurchgangszeit wird eine Information extrahiert, die sich auf eine Änderung der Drehzahl des bürstenlosen Motors 2 bezieht. Daher kann eine Änderung der Drehzahl zweckmäßig bestimmt werden.

According to this embodiment of the present invention described above, the following advantage can be provided in addition to the advantages of the sixth embodiment of the present invention previously described in (1) to (5).

• (7) On the basis of the result of detecting the zero-crossing time, information is extracted that relates to a change in the rotational speed of the brushless motor 2 refers. Therefore, a change in the rotational speed can be suitably determined.

Eighth embodiment

In the sixth embodiment of the present invention, when the zero-crossing time occurs before the enabled period, it is assumed that the time of the start of the enabled period is the zero-crossing time to set a certain time. In this case, however, when the zero-crossing time continuously occurs before the released period, the certain time is continuously delayed from the appropriate time, and this deteriorates the efficiency of controlling the output of the brushless motor 2 ,

Therefore is in an eighth embodiment of the present invention, a leading time, which is referred to as the zero-crossing time can be accepted and applied before the time of onset of a release date. When the zero crossing time before the released period occurs, the zero crossing time at a time between the most advanced time and the time the beginning of the approved duration (including the most advanced Time and time of commencement) to which the difference between it and the zero crossing time is minimized.

As it is in 27A and 27B is shown, the leading time is set to a timing at which the value Cma of a maximum-advance-counter becomes zero. The maximum advance counter starts decrementing from its initial value for which the value of the counter is taken at the zero-crossing time. By making the decrementing speed of the maximum advance counter faster than the decrementing speed of the release start counter, the leading time at which the value of the maximum advance counter becomes zero can be set before the time of commencement of a released period. This leading time is determined as the time at which a predetermined electrical angle after the zero-crossing time according to the specifications of the brushless motor 2 or the like passes.

28 Fig. 15 shows processing for coping with an anomaly that the zero-crossing time occurs before the enabled period. This processing is performed by the drive control circuit 220 for example, repeatedly executed in a predetermined cycle. In 28 will be the same processing as the one in 24 is identified with the same step numbers for simplicity.

If a negative determination is made in S52, that is, it determines is that the zero crossing time occurred before the released period is the processing proceeds to S100. In S100 it is checked if one Output level change time the corresponding one of the comparison signals Uc, Vc, Wc in advance or before the most advanced time (which includes the leading time) occurred or not. This processing can for example according to that accomplished whether the logic level of the corresponding one of the comparison signals Uc, Vc, Wc, when the value of the maximum advance counter becomes zero, is abnormal or not. If it is determined that the change time is before the most leading time, the following processing in S102 run: additionally to a processing for determining an excessive overrun of a zero-crossing timing and processing for replacing the value Cm of the counter the value Cp of the release start counter a processing for accepting that the leading time is the zero-crossing time is to set a specific time. That means, the Cs value of the counter to set a particular time is set so that the following is based on the duration of the previous one Zero crossing time is realized at the zero crossing time, the is adopted at this time. The counter for specifying a specific Time assumes a value corresponding to the time of the current time is required at the given time.

If It is determined in S100 that the change time before the most leading time, the following processing in S104 additionally to a processing for determining an excessive overrun of a zero-crossing timing and processing for replacing the value Cm of the counter the value Cp of the release start counter run: a processing for accepting that the modification time of the corresponding the comparison signals Uc, Vc, Wc the zero-crossing time in duration from the most advanced time to the time of the beginning of a shared one Duration is, thereby setting a specific time. The means the value Cs of the counter to set a particular time is set so that the following is based on the duration of the previous one Zero crossing time is realized at the zero crossing time, the is adopted at this time. The counter for specifying a specific Time assumes a value corresponding to the time of necessary at the given time.

When the processing of S102 or S104 is finished, the processing from S56 to S64 becomes 6 executed.

According to this embodiment of the present invention, the following advantage is afforded. Also, when the zero-crossing time occurs from the time of commencement of a released period, a certain timing can be set by setting the most advanced time using a timing as close as possible to the actual zero-crossing timing. For this reason, deterioration of the efficiency of controlling with respect to the output of the brushless motor 2 be suppressed advantageous. Further, based on that the zero-crossing time continuously occurs before the released period, any abnormality in the rotating state is determined; Therefore, this determination can also be carried out expediently.

The sixth to eighth embodiments of the present invention be configured as described below.

In the sixth embodiment of the present invention, the count speed of the release start counter and the permitted period setting counter is changed according to the change ΔVB of the voltage VB of the battery 214 corrected. As it is in 29 instead, the counting speed may be corrected according to two parameters: the voltage VB of the battery 214 and their change ΔVB. 29 Fig. 16 shows a data map defining the relationship between these two parameters and the count speed correction coefficient Aij. The reason why the voltage VB of the battery 214 when a parameter is used is as follows. Even if the change ΔVB the voltage of the battery 214 is identical, the degree of change in the speed of the brushless motor 2 depending on the voltage VB of the battery 214 differ. For this reason, using two parameters allows rotation fluctuation of the brushless motor 2 caused by a fluctuation of the voltage VB of the battery 214 caused to accurately detect.

The parameters for detecting a change in the speed of the brushless motor 2 are not limited to those described in connection with the preceding embodiments of the present invention and their embodiments. For example, the magnetic flux in the brushless motor 2 attenuated and becomes a speed with an increase in the temperature Tm of the brushless motor 2 elevated. In consideration of this, a change in the temperature (or its equivalent) of the brushless motor 2 be used. Alternatively, by using a map showing the relationship between the count speed correction coefficient Bij and the temperature Tm of the brushless motor 2 and a change ΔVB in the voltage of the battery 214 defines how it is in 30 is shown, a change in the speed can be detected on the basis of these two parameters.

In the sixth embodiment of the present invention, the processing of 22 not be executed. A synchronization loss state may be determined from at least the influence of a rotation fluctuation due to a start or a sudden noise by setting the threshold MAX in 24 or 25 be differentiated.

The zero point voltage of the brushless motor 2 can be used as the reference voltage Vref instead of the virtual zero point voltage. One half of the voltage VB of the battery 214 can by dividing the voltage of the battery 214 be used with a resistive element.

When a PWM control as a method of controlling the brushless motor 2 For example, the following measure can be used. The period for which the switching elements SW1 to SW6 are ON is taken as an ON-release period, and the switching elements SW1 to SW6 are repeatedly turned on and off in this duration. In this case, however, the speed of the brushless motor 2 in accordance with the ratio (operating time) of the ON time to the sum of the ON time and the OFF time. Therefore, it is desirable to determine a mode for correcting the counting speed according to the operating time.

The timing at which a predetermined electrical angle occurs, which is detected based on the induced voltage, is not limited to the zero-crossing timing. For example, the one that can be applied in the JP 11-18478A is detected.

The power supply, with the brushless motor 2 connected, does not have the battery 214 but a generator can be connected. The brushless motor 2 does not have to be an actuator of an on-board fuel pump, but may be an actuator of an in-vehicle cooling fan. The lathe ne need not be a three-phase brushless motor, but may be an engine of any number of phases. Furthermore, he does not have to be an engine, but he can be a generator.

Ninth embodiment

In the ninth embodiment according to the present invention, which is incorporated in 31 is shown is a brushless motor 2 a three-phase motor, and an actuator is a fuel pump for an internal combustion engine installed in a motorcycle. The three phases (U-phase, V-phase, W-phase) of the brushless motor 2 are with an inverter 12 connected. This inverter 12 is a three-phase inverter and sets the voltage VB of a battery 214 to the three phases of the brushless motor 2 at. To connect a line between each of the three phases and the positive pole or negative pole of the battery 214 is the inverter 12 configured to include a parallel circuit body comprising: switching elements SW1, SW2 (U-phase leg), switching elements SW3, SW4 (V-phase leg), and switching elements SW5, SW6 (W-phase leg). The switching elements SW1 to SW6 and diodes D1 to D6 are in the similar manner as in the sixth embodiment of the present invention ( 18 ) connected.

A drive control circuit 230 operates the inverter 12 via a driver 222 and thereby controls the output of the brushless motor 2 , More specifically, the drive control circuit takes 230 Comparison signals PU, PV, PW of comparators 224 . 226 . 228 and operates the switching elements SW1 to SW6 based on these.

The comparators 224 . 226 . 228 serve to compare the terminal voltage Vu, Vv, Vw of respective phases with a reference voltage Vref. The comparator 224 compares the U-phase terminal voltage Vu of the brushless motor 2 with the reference voltage Vref and outputs the result of this comparison as a comparison signal PU. The comparator 226 compares the V-phase terminal voltage Vv of the brushless motor 2 with the reference voltage Vref and outputs the result of this comparison as a comparison signal PV. The comparator 228 compares the W-phase terminal voltage Vw of the brushless motor 2 with the reference voltage Vref and outputs the result of this comparison as a comparison signal PW.

In this embodiment of the present invention, a virtual zero point voltage obtained by dividing the terminal voltages Vu, Vv, Vw of the respective phase with resistive elements RU, RV, RW is used as the reference voltage Vref. The reason is as follows. In the vehicle's own battery 214 Its voltage tends to fluctuate rapidly while the rate of change of induced voltage in the brushless motor 2 tends to be less than the rate of change of voltage of the battery 214 to become. For this reason, when the voltage of the battery 214 rises rapidly, 1/2 of the amplitude of an induced voltage not equal to 1/2 the voltage of the battery 214 , For this reason, if 1/2 the voltage of the battery 214 as the reference voltage is used, a comparison by the comparators 224 . 226 . 228 not appropriate with the fluctuating voltage of the battery 214 be executed.

The drive control circuit 230 switches the switching elements SW1 to SW6 through the driver 222 in and out. In this example, it mainly executes switching control by 120 ° excitation method. Using the previous comparison signals PU, PV, PW, it detects a time (zero crossing time) at which the induced voltage of each phase of the brushless motor 2 equal to the reference voltage Vref of the brushless motor 2 becomes. Then, it changes the operation of the switching elements SW1 to SW6 at a timing (specific time) delayed by a predetermined electrical angle (for example, 30 °) from the zero-crossing timing.

The drive control device 230 may be constructed by logic circuits or may be constructed by a central processing unit and a memory unit for storing a program.

The switching control is carried out in a 120 ° excitation control as shown in FIG 32 is shown. More specifically, (a) represents the transition of the terminal voltages Vu, Vv, Vw indicated by solid lines and the reference voltage Vref by dotted lines. In this embodiment of the present invention, the virtual neutral voltage is used as the reference voltage Vref. Although, in fact, the reference voltage Vref varies, it is considered to be constant here for the sake of simplicity. (b) illustrates the results of a comparison of the terminal voltages Vu, Vv, Vw with the reference voltage Vref with respect to amplitudes (comparison signals PU, PV, PW). (c) represents the transition of a logically combined signal PS of the comparison signals PU, PV, PW , 2 (d) put the over transition of a logically combined signal (expectation signal Se) of the comparison signals PU, PV, PW, which are expected when the zero-crossing time occurs while the switching elements SW1 to SW6 are operated. 2 (e) represents the transition of a detection signal Qs with respect to the zero-crossing timing. This is a signal whose rising edges and falling edges are synchronous with the zero-crossing timing. (f) represents the transition of the values of various counters, and (g) represents the transition of activation signals for the switching elements SW1 to SW6. The activation signals shown in (g) include activation signals U +, V +, W + for the switching elements SW1, SW3, SW5 of a high side of the branches of the respective phases and enable signals U-, V-, W- for the switching elements SW2, SW4, SW6 of a low side of the branches of the respective phases. The high side switch elements SW1, SW3, SW5 of the respective phases are P-channel transistors; therefore, the durations for which these activation signals U +, V +, W + are at a logical L are the durations for which they are ON.

The combined signal PS is a three-bit signal, and the respective logic values of the comparison signals PU, PV, PW coincide with the logic values of its highest-order (highest) bit, intermediate-bit, and lowest-order (low) bits. That is, when the comparison signal PU is at logical H, the highest order bit is set to 1; and when the comparison signal PU is at logical L, the highest order bit is set to 0. For this reason, when the comparison signals PU, PV, PW are respectively at H, L, and H, for example, the combined signal PS is on 101 in a binary notation and 5 in a decimal notation. In 32 both the combined signal PS and the expectation signal Se are shown in decimal notation.

The solid line in (f) shows the value Cm of a measuring counter for Measuring a time interval between adjacent zero crossing times at. The meter is initialized at any time, at which time the zero crossing time occurs and starts a time counting process New. A time interval between adjacent zero-crossing times has a correlation to a speed. That's why the value of the counter, immediately before being initialized (the maximum value of the counter) one Parameter, which has a correlation to a rotational speed.

The dotted lines in (f) show the value Cs of a counter for Set a counter to set a specific time that counts a time that is required until a zero crossing time equal to a certain Time, thereby defining a particular point in time. The counter to Setting a specific time takes the value of the counter an initialization as its initial value at the zero-crossing time on and decrements this. Then he sets the time to which the Value zero is fixed as a specific time. At this time the following procedure is performed. If the time interval between the zero crossing time and the specific time for example 30 °, For example, the decrementing speed becomes twice the incrementing speed of the measuring counter established. Considering its that the time interval between adjacent zero crossing times 60 °, it may be considered that this setting makes it possible the time at which the value of the counter sets a particular Time zero becomes 30 degrees from to delay the zero crossing time.

The two-dot chain lines shown in (f) indicate the value Cmk of a masking duration counter. The masking duration counter determines a duration (masking duration), for which detects the zero-crossing time based on the Comparison of the terminal voltages Vu, Vv, Vw with the reference voltage Vref suppressed (blocked). The counter serves to avoid the following event. When the terminal voltages Vu, Vv, Vw coincide with the reference voltage Vref during a period, for which a stream over the diodes D1 to D6 supplied is, the zero crossing time is detected incorrectly. This counter takes as well as the value of the meter before initialization as its initial value at the zero-crossing time on and decrements this. Then he sets the duration before the value Zero is fixed as a masking period. When the masking time for a duration from the zero-crossing time to, for example, 45 ° For example, the decrement rate may be on 3/2 times the Incrementing rate of the measuring counter.

If the value of the masking duration counter Zero is, the comparison signals are PU, PV, PW and the combined Signal PS released. If the combined signal PS with the expected signal Se while this duration matches, the detection signal Qs is inverted. The zero crossing time starts when the detection signal Qs is inverted, the counter for specifying a certain one Time a decrement. If its value is zero, the Operation of switching elements SW1 to SW6 changed.

As it is in 32 is shown, have the specific time at which the switching elements SW1 to SW6 and the zero-crossing timing are one-to-one correspondence with each other. For this reason, the behavior of the terminal voltages Vu, Vv, Vw of the respective phases is determined uniquely in accordance with the operating state of the switching elements SW1 to SW6. Consequently, the previous expectation signal Se can be uniquely determined.

When the battery 214 and the inverter 12 or any other similar event occurs, the following phenomenon may take place. Due to transmitting a vibration of the vehicle to the battery 214 and any other reason can the battery 214 and the inverter 12 are immediately separated from each other and then a re-routing is formed between them. When a power supply to the brushless motor 2 is temporarily interrupted at this time, the speed of the brushless motor 2 reduced. When a fuel that is discharged from a fuel tank to the upstream side by a fuel pump, that of the engine 2 is driven back at this time, a force is applied to the opposite rotation side on the brushless motor 2 exercised. This can eventually cause the engine 2 turns backwards. In this situation, if the switching elements SW1 to SW6 are operated under normal conditions, a problem may arise. For example, an oscillation phenomenon occurs that the brushless motor 2 a normal rotation and a reverse rotation is repeated, and it is difficult to use the brushless motor 2 to control in a convenient rotational state.

The reverse rotation of the brushless motor 2 may be appropriately detected on the basis of the previous combined signal PS consisting of three bits. As it is in 33 should be shown, if the brushless motor 2 rotates normally in the forward direction, time series data relating to the combined signal PS coincide with time sequence data with respect to the expectation signal Se. If the brushless motor 2 Meanwhile, time-series data regarding the combined signal PS should be coincident with data obtained by reversing in time, that is, reversing the order of arrangement, time-series data with respect to the expectation signal Se, as shown in FIG 34 is shown. For this reason, the reverse rotation of the brushless motor 2 be detected on the basis of the combined signal PS.

When the rotational state of a brushless motor 2 is abnormal, all of the switching elements SW1 to SW6 are turned off and the operation is awaiting the brushless motor 2 to stop. Then the brushless motor 2 restarted. In this case, however, it takes a long time to get the brushless motor 2 bring back to a normal state.

To cope with this, the following processing is carried out in this embodiment of the present invention. If it is detected that the brushless motor 2 Turning backwards, processing is performed to stop the rotation of the brushless motor 2 forcibly stopping, and then a restart processing is executed.

35 Provides processing to restart the brushless motor 2 in this embodiment of the present invention. This processing is performed by the drive control circuit 230 for example, repeatedly executed in a predetermined cycle.

This series of processing is performed as follows. In S210, it is checked if the value Cmk of the masking duration counter is zero or not. If it is determined that the value Cmk of the masking duration counter is zero, it is determined in S212 whether or not the combined signal PS of the comparison signals PU, PV, PW has been changed. This processing is for determining whether or not it is the zero-crossing timing. If it is determined that the combined signal PS has been changed, it is determined in S214 whether or not the current combined signal coincides with the second preceding expected signal Se. This processing is used to determine if the brushless motor 2 turns backwards or not. As it is in 33 is shown, when the brushless motor 2 reverse rotation, time-series data with respect to the combined signal PS inverted. Therefore, it is assumed that the current combined signal PS coincides with the second preceding expectation signal Se. If the current combined signal PS coincides with the second preceding expectation signal Se, it is determined in S16 that the brushless motor 2 turns backwards.

In S218, processing is subsequently performed to control the rotation of the brushless motor 2 forcibly stop. More specifically, the switching elements SW1, SW3, SW5 or the switching elements SW2, SW4, SW6 are all turned on to all of the phases of the brushless motor 2 to close briefly. Therefore, a current through the brushless motor 2 merely supplied by an induced voltage, the in connection with the rotation of the brushless motor 2 is produced. This current is rapidly attenuated by the resistance of the current passage. As a result, the rotational energy of the brushless motor 2 converted into an electric energy and then weakened. For this reason, the brushless motor 2 be stopped quickly.

In S220 it checks if the speed of the brushless motor 2 is essentially zero or not. The speed may be calculated based on time intervals between adjacent zero crossing times. This can be done using the maximum value Cm of the counter. When it is determined that the rotational speed is substantially zero, restart processing in S222 is performed by operating the switching elements SW1 to SW6 based on the zero-crossing timing.

If in any of S210 to S214, a negative determination is made or when the processing of S222 is finished, this row is finished processing once.

Therefore, the zero-crossing timing is used in determining whether or not the rotational speed associated with a stop operation is substantially zero. However, this can cause a problem. The virtual zero point is used to set the reference voltage Vref. Therefore, when the speed of the brushless motor 2 is substantially zero, the terminal voltages Vv, Vu, Vw of all the phases of the brushless motor 2 match the reference voltage Vref. When all the switching elements SW1 to SW6 are turned off once to restart the engine, all the phases are brought to a high impedance state. Therefore, also in this case, the terminal voltages Vv, Vu, Vw of all the phases of the brushless motor 2 match the reference voltage Vref.

When noise is superimposed on the terminal voltages Vv, Vu, Vw in these situations, the following phenomenon occurs. The terminal voltages Vv, Vu, Vw, which are indicated by solid lines in 36A and the reference voltage Vref, represented by single-dotted lines, cross each other frequently. In fact, the reference voltage Vref also varies in accordance with the terminal voltages Vu, Vv, Vw; however, it is considered to be constant here for the sake of simplicity. In this case, there is a possibility that the zero-crossing timing is often erroneously detected. In this embodiment of the present invention, to cope with this, the reference voltage is subjected to offset correction, so that the following is realized. The value of the reference voltage Vref and the values of the terminal voltages Vv, Vu, Vw applied to the comparators 224 . 226 . 228 be entered as it is in 31 are made different from each other when the rotational speed is substantially zero. That is, the signal wire for inputting the reference voltage Vref to the comparators 224 . 226 . 228 is about a resistive element 30 with the positive pole of the battery 214 connected. Therefore, the following can be realized as shown in (b) of 35 is shown. The reference voltage Vref obtained when the rotational speed is substantially zero can be offset to the positive pole side of the battery by an offset Δ with respect to the terminal voltage Vu 214 Getting corrected. In this example, only the U phase is shown.

This offset Δ is set to such a small value that controlling with respect to the brushless motor 2 is not affected. At the same time, the offset is set to a value which is assumed to be required to realize the following. Even when noise is mixed, the terminal voltage Vu and the reference voltage Vref are prevented from crossing each other when the rotational speed is substantially zero as shown in FIG 36B is shown. In this setting, as it is in 36C is shown increases the reference voltage Vref with increasing the speed. When the terminal voltage Vu rises and the reference voltage exceeds Vref, the zero-crossing timing occurs at which the terminal voltage Vu and the reference voltage Vref cross each other. Therefore, as it is in 36D with respect to the U-phase, the control under normal conditions, which in 32 are shown executed.

In particular, in this embodiment of the present invention, a deterioration in the efficiency of the brushless motor 2 by correcting the reference voltage Vref. Meanwhile, when the terminal voltages Vu, Vv, Vw are corrected, the phenomenon occurring in 36E is shown. That is, the value of a fluctuation of the voltage applied to the brushless motor 2 is lower than the voltage VB between the positive pole and the negative pole of the battery 214 and this deteriorates efficiency. That is, attenuation is generated with a value equivalent to the range in 36E hatched.

It is desirable that the offset Δ should be about an order of magnitude smaller than the value VE of a voltage drop observed when a current is supplied through the diodes D1 to D6. This makes it possible to prevent the values of the comparison signals PU, PV, PW from becoming abnormal due to ringing noise when the operation of the switching elements SW1 to SW6 is changed. For this reason, the following can be realized even when PWM control is executed. That is, the following can be realized even if, for example, a duration of a certain time up to 120 ° is taken as a duration of a released ON operation and the switching elements SW1 to SW6 are turned on and off during this period. A zero crossing time can be determined by the method described in 32 is shown accurately.

According to the ninth embodiment The present invention provides the following advantages.

• (1) When the speed of the brushless motor 2 is substantially zero, the terminal voltages Vu, Vv, Vw become equal to each other. At this time, at least one of the values of the terminal voltages Vu, Vv, Vw and the value of the reference voltage Vref, which are from the comparators 224 . 226 . 228 are subjected to an offset correction. These values are thereby made different from each other. Therefore, even when noise is mixed and therefore a terminal voltage fluctuates, occurrence of a phenomenon that the terminal voltage frequently crosses the reference voltage can be avoided. As a result, erroneous detection of the zero-crossing timing can be advantageously avoided.
• (2) The previous correction is made by the resistive element 30 executed, the signal wire for inputting the reference voltage Vref in the comparators 224 . 226 . 228 with the positive potential of the battery 214 combines. Therefore, the reference voltage Vref may become the positive potential of the battery 214 can be corrected and therefore the value of a correction by the resistance value of the resistive element 30 be set.
• (4) Since the reference voltage Vref is to be corrected, the following advantage is produced. A deterioration of the efficiency of controlling the brushless motor 2 can be suppressed compared with cases where the terminal voltages Vu, Vv, Vw are to be corrected, and further, the number of objects to be corrected can be reduced.
• (5) The virtual zero point voltage is used for the reference voltage Vref. Therefore, even if the brushless motor 2 has no wiring connected to a zero point, the reference voltage Vref is appropriately determined.
• (6) The operation of the switching elements SW1 to SW6 for starting the brushless motor 2 at a stop is started based on a detected value of a zero crossing time point. In this case, when noise is mixed to the terminal voltages Vu, Vv, Vw, a phenomenon that the terminal voltages Vu, Vv, Vw and the reference voltage Vref frequently cross each other may occur. As a result, erroneous detection of the zero-crossing timing is prone to occur. To cope with this, this embodiment of the present invention is constructed such that the foregoing operation and advantage can be particularly advantageously provided.

In the ninth embodiment of the present invention, microcomputer processing may be used instead of the comparators 224 . 226 . 228 are used to compare the terminal voltages Vu, Vv, Vw and the reference voltage Vref. The signal wire for applying the reference voltage Vref to the comparators 224 . 226 . 228 can be grounded via a resistive element. Therefore, the reference voltage Vref can be corrected to the ground potential side. The structure for correcting the reference voltage Vref need not be such that the signal wire for applying the reference voltage Vref to the comparators 224 . 226 . 228 is connected via a resistive element with a predetermined potential. For example, an output signal caused by voltage change of the positive potential of the battery 214 is achieved via an inverting amplifier circuit or a non-inverting amplifier circuit, to the signal wire for applying the reference voltage Vref to the comparators 224 . 226 . 228 be created. The object to be corrected need not be the reference voltage Vref and may be the terminal voltages Vu, Vv, Vw. Alternatively, it may be both the reference voltage Vref and the terminal voltages Vu, Vv, Vw. However, when the terminal voltages Vu, Vv, Vw are corrected, it is desirable to make their correction values identical with each other.

Tenth embodiment

In the tenth embodiment of the present invention, the values of the terminal voltages Vu, Vv, Vw and the value of the reference voltage Vref to be compared with each other are transmitted setting the comparators 224 . 226 . 228 made different from each other. Each of the comparators 224 . 226 . 228 is built as it is in 37 is shown.

Each comparator is constructed to include a differential amplifier circuit 240 and an output circuit 250 includes. In the differential amplifier circuit 240 is a constant current source 241 that with the positive pole of the battery 214 connected to the emitters of a pair of transistors 242 . 243 connected. The bases of the pair of transistors 242 . 243 are connected to the non-inverting input terminal (+) or the inverting input terminal (-) of the comparator 224 . 226 . 228 connected. The collectors of the transistors 242 . 243 are with the collectors of the transistors 244 respectively. 245 connected. The bases of the transistors 244 . 245 are shorted to each other and the emitters of the transistors 244 . 245 are over resistive elements 246 . 247 grounded. The bases of the transistors 244 . 245 are connected to the connection point between the transistor 242 and the transistor 244 connected.

The output circuit 250 includes a transistor 252 and a transistor 254 and the base of the transistor 252 is with the connection point between the transistor 243 and the transistor 245 connected. The emitter of the transistor 252 is grounded and its collector is with the constant current source 241 connected. The base of the transistor 254 is with the collector of the transistor 252 connected and the emitter of the transistor 254 is grounded. The collector of the transistor 254 is having a power supply by grading the positive potential of the battery 214 is achieved to a predetermined voltage, via a resistive element 256 connected. Furthermore, it forms the output terminal of the comparator 224 . 226 . 228 ,

In this structure, the following operation is performed. When the voltage of the non-inverting input terminal is higher than the voltage of the inverting input terminal, the transistor becomes 243 turned on and becomes the transistor 242 switched off. Therefore, a current flows from the constant current source 241 over the transistor 243 to the base of the transistor 252 , As a result, the transistor 252 turned on and becomes the transistor 254 switched off. Therefore, the output signal of the comparator 224 . 226 . 228 to the positive potential of the battery 214 brought. In other words, the comparator outputs a logical H signal.

When the voltage of the non-inverting input terminal is lower than the voltage of the inverting input terminal, the transistor becomes 242 turned on and becomes the transistor 243 switched off. As a result, a current flows from the constant current source 241 over the transistor 242 into the bases of the transistors 244 . 245 , Because of this, the transistors become 244 . 245 turned on and no current flows to the base of the transistor 252 , Because of this, the transistor becomes 252 turned off and becomes the transistor 254 switched on. Therefore, the output signal of the comparator 224 . 226 . 228 brought to ground potential. In other words, the comparator outputs a signal of logic L.

The accuracy of a comparison of a pair of input signals in amplitude by the differential amplifier circuit 240 depends on the symmetry of pairs of elements in the differential amplifier circuit 240 which correspond to the input terminals. The pairs of elements include the transistor 242 and the transistor 243 , the transistor 244 and the transistor 245 and the resistive elements 246 and the resistive element 247 , More specifically, the accuracy of a comparison is increased with increasing symmetry.

In this embodiment of the present invention, a pair of elements corresponding to the input terminals are made asymmetrical. As a result, when the values of voltages applied to the non-inverting input terminal and the inverting input terminal are identical, the output of the comparator increases 224 . 226 . 228 without failure a certain logical value. This is because in the differential amplifier circuit 240 any of the transistors 242 . 243 even if an identical voltage is applied to the non-inverting input terminal and the inverting input terminal, it tends to be turned on.

When the transistors 242 . 243 are constructed so that they are asymmetrical, for example, the following is realized. When the height UVF + a voltage drop between emitter and collector due to that the transistor 242 is turned on, greater than the height UVF- between emitter and collector due to which is that the transistor 243 is turned on, the transistor tends 243 more to be turned on. When the transistors 244 . 245 are constructed so that they are asymmetric, the following is realized. When the height UVf + a voltage drop between emitter and collector due to that the transistor 244 is turned on, greater than the height UVf- a voltage drop between the emitter and the collector because of this is that the transistor 245 is turned on, the transistor tends 243 more to be turned on. When the resistance of the resistive element 246 higher than the resistance value of the resistive element 247 is, the transistor tends 243 more to be turned on.

With this setting, the following is realized. In a situation where the reference voltage Vref and the terminal voltages Vu, Vv, Vw coincide with each other, the comparators can be made to be 224 . 226 . 228 determine without failure that any one is higher. For this reason, it may also be in the situation in 36A is shown, the output values of the comparators are avoided 224 . 226 . 228 frequently inverted. In order to accurately bring an interval between zero-crossing times corresponding to an interval of 60 °, it is desirable to perform the following measure. In a situation where the true value of the reference voltage Vref and the true values of the terminal voltages Vu, Vv, Vw coincide with each other, the results of the comparison by the comparators become 224 . 226 . 228 with respect to whichever is higher, the terminal voltages Vu, Vv, Vw or the terminal voltages are made identical to each other. Furthermore, it is desirable to have the difference between the reference voltage Vref and the terminal voltage Vu, Vv, Vw when the output of the comparator 224 . 226 . 228 is inverted for all of the phases to be identical. In this embodiment of the present invention, for this purpose, the comparators 224 . 226 . 228 made identical in structure to each other.

• (7) Paare von Elementen in dem Differentialverstärkerschaltungen 240, die jeweils den Paaren von Eingangsanschlüssen der Komparatoren 224, 226, 228 entsprechen, sind derart aufgebaut, dass sie asymmetrisch sind. Daher kann die relative Amplitudenbeziehung zwischen einem Paar von Eingangssignalen, das miteinander zu vergleichen ist, verschoben werden. Aus diesem Grund kann das Folgende in einer derartigen Situation realisiert werden, dass die Drehzahl des bürstenlosen Motors 2 im wesentlichen Null ist und daher die Referenzspannung Vref und die Anschlussspannungen Vu, Vv, Vw zueinander gleich werden. Die relative Differenz zwischen diesen Werten, die miteinander zu vergleichen sind, kann erweitert werden.
• (8) Der Aufbau der Komparatoren 224, 226, 228 wird für alle der Phasen identisch gemacht. Daher können die Intervalle zwischen Nulldurchgangszeiten genau auf 60° festgelegt werden.

According to this embodiment of the present invention, the following advantages can be provided in addition to the advantages of the ninth embodiment of the present invention.

• (7) Pairs of elements in the differential amplifier circuits 240 , respectively, the pairs of input terminals of the comparators 224 . 226 . 228 are constructed to be asymmetrical. Therefore, the relative amplitude relationship between a pair of input signals to be compared with each other can be shifted. For this reason, the following can be realized in such a situation that the rotational speed of the brushless motor 2 is substantially zero and therefore the reference voltage Vref and the terminal voltages Vu, Vv, Vw become equal to each other. The relative difference between these values, which are to be compared, can be extended.
• (8) The construction of the comparators 224 . 226 . 228 is made identical for all phases. Therefore, the intervals between zero crossing times can be set exactly at 60 °.

In the tenth embodiment of the present invention, the circuit of the comparators 224 . 226 . 228 be constructed so that they include MOS transistors.

In both of the ninth and tenth embodiments of the present invention Invention can the following treatments are performed.

The counter for setting a specific time and the measuring counter can be made identical in the counting speed, and the initial value of the counter for setting a specific time is set according to the value of the counter before initializing (maximum value). For example, when the specific time is set to a time delayed by 30 ° from the zero-crossing time, 1/2 of the maximum value of the measurement counter is taken as the initial value of the counter to set a specific time. The masking duration counter and the measuring counter can be made identical in a counting speed, and the initial value of the masking duration counter is set according to the value of the counter before initializing (maximum value). When an angle range from the zero crossing time of 45 ° is set as the masking period, 3/4 of the maximum value of the counter is taken as the initial value of the masking duration counter. Instead of the virtual zero point voltage, the zero point voltage of the brushless motor 2 be used for the reference voltage Vref. The switching elements SW1, SW3, SW5 on the high side of the respective branches may be constructed of an N-channel MOS transistor. The power supply, with the brushless motor 2 connected, does not have the battery 214 but can be a generator. The brushless motor 2 does not have to be an actuator of an on-board fuel pump, but may be an actuator of an in-vehicle cooling fan. The lathe need not be a three-phase brushless motor, but may be an engine of any number of phases. Furthermore, it does not have to be an engine, but can be a generator.

Eleventh embodiment

In the eleventh embodiment of the present invention, a lathe driving device as shown in FIG 38 is shown in the similar manner as in the sixth embodiment of the present invention ( 18 ) and the ninth embodiment of the present invention ( 31 ) built up. However, in this embodiment of the present invention, there is a current detecting device 228 instead of the voltage detection device 225 ( 18 ) intended. Therefore, currents conducted through the switching elements SW1 to SW6 are detected based on their resistance values. That is, the current detection device 228 includes sense transistors for detecting the currents respectively conducted through the switching elements SW1 to SW6. The sources of the switching elements SW1 to SW6 and the corresponding detection transistors are short-circuited together and their gates are short-circuited together, and thereby current mirror circuits are constructed. Therefore, the currents conducted by the switching elements SW1 to SW6 can be detected based on the output currents of the detection transistors. In fact, it is desirable that the current sensing device 228 near the inverter 12 should be trained. For information, a method of constructing a current mirror circuit to detect a current, for example, in US Pat JP-A-10-256541 disclosed.

The switching control device 227 switches the switching elements SW1 to SW6 through the driver 222 in and out. In this example, it mainly performs switching control by a 120 ° excitation method. More specifically, a virtual zero point voltage (reference voltage Vref) becomes as a result of voltage division by resistive elements RU, RV, RW with respect to the terminal voltages Vu, Vv, Vw of the respective phases of the brushless motor 2 achieved. Based on a timing at which this virtual zero point voltage matches the terminal voltage Vu, Vv, Vw of each phase of the brushless motor 2 is coincident, the time (zero-crossing point) at which an induced voltage agrees with the reference voltage Vref is detected. Then, it changes the operation of the switching elements SW1 to SW6 at a timing (specific time) delayed by a predetermined electrical angle (for example, 30 °) from the zero-crossing timing. When the current from the current-sensing device 228 However, the following measure is taken to limit the current (amount of excitation) caused by the brushless motor 2 is directed. Instead of taking a duration of 120 ° as a duration for which the switching elements SW2, SW4, SW6 are turned on, PWM control is executed during this period.

The switching control device 227 may be constructed as a logic circuit or may be constructed as a central processing unit and a memory unit for storing a program.

39 FIG. 12 illustrates the manner in which a shift control by the shift control means 227 is performed in a 120 ° excitation control. More specifically, (a) represents the transition of the terminal voltages Vu, Vv, Vw shown by solid lines and the reference voltage Vref. In this embodiment of the present invention, the virtual neutral point voltage is used for the reference voltage Vref. Although, in fact, the reference voltage Vref varies, it is considered to be constant here for the sake of simplicity. (b) represents the transition of the results of a comparison of the terminal voltages Vu, Vv, Vw to the reference voltage Vref for an amplitude (comparison signals PU, PV, PW). (c) represents the transition of a logically combined signal PS of the comparison signals PU, PV, PW. (d) represents the transition of a logically combined signal (expectation signal) Se of the comparison signals PU, PV, PW which are expected when the zero-crossing time occurs while the switching elements SW1 to SW6 are operated. (e) represents the transition of a detection signal Qs with respect to the zero-crossing timing. This is a signal whose rising edges and falling edges are synchronized at a zero-crossing time. (f) represents the transition of the values of various counters, and (g) represents the transition of activation signals for the switching elements SW1 to SW6. The activation signals shown in (g) include activation signals U +, V +, W + for the switching elements SW1, SW3, SW5 of a high side of the branches of the respective phases and enable signals U-, V-, W- for the switching elements SW2, SW4, SW6 of a low side of the branches of the respective phases. The high side switch elements SW1, SW3, SW5 of the respective phases are P-channel transistors; therefore, the durations for which activation signals U +, V +, W + are at logic L are the durations for which they are turned on.

The combined signal PS is a three-bit signal and the respective logical values of the comparison signals PU, PV, PW coincide with the logical values of its most significant bit, intermediate bits and least significant bits. That is, when the comparison signal PU is at logical H, the highest order bit is set to 1; and when the comparison signal PU is at logical L, the most significant bit is set to 0. For this reason, when the comparison signals PU, PV, PW are at H, L and H, for example, the combined signal PS is in a binary notation 101 and set to 5 in decimal notation. In 39 Both the combined signal PS and the expectation signal are shown in a decimal notation.

The solid lines in (f) represent the value Cm of a measurement counter for measurement a time interval between adjacent zero crossing times As shown, the count value Cm of the measurement counter becomes initialized at each time the zero crossing time occurs, and restarts a time counting process. A time interval between adjacent zero crossing times points a correlation to a speed. That's why he sees Value Cm of the counter, immediately before being initialized (the maximum value of the counter) one Parameter, which has a correlation to a rotational speed.

The a single dot-dash line in (f) sets the value Cs of a counter Set a specific time that counts a time from then on, when the zero crossing time occurs, until then when the specific time occurs, is required, and lays down a certain time. The counter for specifying a specific Time takes the value of the counter before initializing as its initial value at the zero-crossing time on and decrements it. Then he sets the time to which the value zero is set as a certain point in time. At this time, the following operation is performed. If the interval between the zero crossing time and the determined one Time is for example 30 °, the decrementing speed becomes twice the incrementing speed of the measuring counter established. Considering its that the time interval between adjacent zero crossing times 60 °, it may be considered that it allows this setting, the time at which the value of the counter sets a particular Time zero becomes 30 ° from the Delay zero crossing time.

The double-dashed lines in (f) represent the value Cmk of a Masking period counter The masking duration counter determines a masking duration for which detecting the zero-crossing time based on the comparison of the terminal voltages Vu, Vv, Vw with the reference voltage Vref with respect to an amplitude suppressed (disabled) becomes. This counter serves to prevent the following event. When the terminal voltages Vu, Vv, Vw coincide with the reference voltage Vref during a period, for which a stream over the diodes D1 to D6 supplied is, the zero crossing time is detected incorrectly. This counter takes as well as the value of the meter before initializing as its initial value at the zero-crossing time on and decrements it. Then he sets the duration before the value is set to zero as a masking period. When the masking time is set to a duration of, for example, zero-crossing time of 45 ° the decrementing rate can be 3/2 times the Incrementing speed of the measuring counter are made.

If the value of the masking duration counter is made zero, the comparison signals PU, PV, PW and the combined signal PS allowed. If the combined signal PS with the expectation signal SE during this duration matches, the detection signal Qs is inverted. At the zero crossing time When the detection signal Qs is inverted, the counter starts for Set a specific time a decrement, and if its value is made zero, the operation of the switching elements SW1 changed to SW6.

Of the certain time at which the switching elements SW1 to SW6 turned on and the zero crossing time point have a one-to-one correspondence with each other on. For this reason, the behavior of the terminal voltages Vu, Vv, Vw of the respective phases clearly in accordance with the operating state the switching elements SW1 to SW6 determined. Therefore, the previous one Expectant signal Se are uniquely determined.

Next, processing for a 120 ° excitation control is carried out as shown in FIG 40 and 41 is shown. The processing for setting the counter values with respect to the previous three counters is repeated by the drive control circuit 220 for example, executed in a predetermined cycle.

In this series of processing, it is checked in S310 whether the value Cmk of the masking duration counter is zero or not. If it is determined that the value is zero, it is determined in S312 whether or not the combined signal PS of the comparison signals PU, PV, PW has been changed. If it is determined in S312 that the combined signal PS has been changed, it is determined in S314 whether or not the combined signal PS and the expected signal Se coincide with each other. This processing is to determine whether or not a change in the amplitude relationship between the terminal voltages Vu, Vv, Vw and the reference voltage Vref coincides with a change assumed by the operating state of the switching elements SW1 to SW6. When it is determined that the combined signal PS and the expectation signal Se coincide with each other, it is checked whether an inversion enable flag Fip has been set to ON or not. The inversion release flag Fip is a flag, which is set to ON when the detection signal Qs has not been inverted yet after the mask length counter value Cmk has been made zero. For this reason, when the combined signal PS and the expectation signal Se coincide with each other for the first time after the value of the masking duration counter has been made zero, the inversion enable flag Fip is set to ON.

If the inversion release flag has been set to ON the detection signal Qs is inverted in S318. In S320, the inversion release flag becomes set to OFF. In S322, the value Cm of the measuring counter is subsequently referred to as the values Cs and Cmk re of the meter for setting a specific time and the masking duration counter. In S324 the measuring counter becomes initialized (Cm = 0).

If In S310, a negative determination is made, the value Cm of the measuring counter incremented in S326. In S328 it is checked below if the Cs value of the counter to set a specific time is zero or not. If the value to set a specific time is zero the previous inversion enable flag is set to ON in S330. If the Cs of the counter To set a specific time is not zero, the Cs value of the counter decremented to set a particular time in S332.

If processing S330 or S332 is finished, it is determined in S334, whether the Cmk of the masking duration counter Zero or not. If the value of the masking duration counter is not Is zero, the masking duration counter is decremented in S336.

If in S334 an affirmative determination is made, if in any from S312 to S316, a negative determination is made and when the processing of S336 is finished, this row is finished processing once.

The processing of 41 is for changing the operation of the switching elements SW1 to SW6 on the basis of the previous counter to set a certain time in a 120 ° excitation control. This processing is performed by the drive control circuit 220 for example, repeatedly executed in a predetermined cycle.

In this processing, it is checked in S340 whether or not the value Cs of the counter for setting a specific time has been made zero. This processing is for determining whether or not it is the time for changing the operation of the switching elements SW1 to SW6. When it is determined that the value of the counter for setting a specific time has been made zero, the operation of the switching elements SW1 to SW6 is changed in S342. The operation of the switching elements is changed on the basis of an operation pattern (switching pattern) of the switching elements SW1 to SW6. More specifically, although the operation pattern of the switching elements SW1 to SW6 is changed at intervals of an electrical angle of 60 °, as shown in FIG 39 is shown, this has a periodicity of 360 °. For this reason, the next operating state of the switching elements SW1 to SW6 is uniquely determined from the present operating state. Consequently, the operation of the switching elements SW1 to SW6 is changed on the basis of this unique relationship.

In S344, the expectation signal Se is subsequently updated. If the operating state of the switching elements SW1 to SW6 changes it is assumed that a zero crossing time occurs during a period for which this operating condition is maintained. The values of the comparison signals PU, PV, PW at this zero crossing time are clearly out of the Operating state determined. For this reason, the expectation signal Se is updated to a value corresponding to the present operating state equivalent. More specifically, the following processing is performed. If the previous expectation signal is 1, becomes the present expectation signal set to 5; if the previous expectation signal is 5, the present expectation signal is set to 4; if that preceding expectation signal 4, becomes the present expectation signal set to 6; if the previous expectation signal is 6, the present expectation signal is set to 2; if that preceding expectation signal 2 becomes the present expectation signal set to 3; and if the previous expectation signal 3 is, the present expectation signal is set to 1.

If in S340 a negative determination is made and if the processing Finished in S444, this series of processing becomes one time completed.

According to the previous one Processing can 120 ° excitation taxes be carried out expediently.

42 represents processing in the previous PWM control. This processing is performed by the drive control circuit 220 for example, repeatedly executed in a predetermined cycle.

In this processing, it becomes S350 of current sensing device 228 , in the 38 1, determines whether the maximum value of currents Imax passing through the individual phases of the brushless motor 2 go, exceeding a threshold Iref or not. This threshold value Iref can be set, for example, based on the maximum value of currents allowed in the switching elements SW1 to SW6. If it is determined that the threshold is exceeded, PWM processing is executed in S352. In this processing, the switching elements SW2, SW4, SW6 become a low side of the branches of the inverter 12 repeatedly turned on and off during an ON period (duration of ON allowance) determined by the previous predetermined time. If a negative determination is made in S350 and the processing of S352 is finished, this series of processing is completed once.

When the foregoing PWM control is executed, the terminal voltages Vu, Vv, Vw change frequently. According to the processing, which in 40 is shown, however, the zero crossing time can be detected accurately even in this case. 43 FIG. 14 illustrates the manner in which switching control in the PWM control is executed (a) to (e) in FIG 43 correspond to (a) to (e) in 39 ,

This figure illustrates the manner that performs the PWM control during a duration of allowing an ON operation for the switching element SW2 of a low side of the U-phase leg (a duration for which is in a 120 ° excitation control). As illustrated, the U-phase terminal voltage Vu rises and becomes higher than the positive voltage VB of the battery at any time when the switching element SW2 is switched from the ON state to the OFF state 214 , This is because, when the switching element SW2 is switched from the ON state to the OFF state, a voltage that will hold that the current supplied to the U phase when it was ON is conducted , from the inductance component of the brushless motor 2 is produced. At this time, the switching elements SW1, SW2 of the U phase are both OFF; therefore, a current is supplied through the U phase through the diode D1. For this reason, the U-phase terminal voltage Vu becomes higher than the positive voltage VB of the battery 214 about a height equivalent to a voltage drop in the diode D1.

Since the switching elements SW5, SW6 are the W phase both OFF at this time, the W phase is brought to a high impedance state. The W-phase terminal voltage Vw at this time is made equal to the positive voltage VB of the battery by the V-phase terminal voltage Vv and the U-phase terminal voltage Vu, which as a result of the switching element SW3 being turned on 214 have been raised. Therefore, it becomes higher than the positive voltage VB of the battery 214 , For this reason, the reference voltage Vref set from the virtual zero point also becomes higher than the positive voltage VB of the battery 214 at any time when the switching element SW2 is turned off. Although the reference voltage Vref is lower than the U-phase terminal voltage Vu when the switching element SW2 is turned off, it is higher than the W-phase terminal voltage Vw at this time. For this reason, the W-phase terminal voltage Vw becomes lower than holds the reference voltage Vref until the induced voltage of the W phase becomes equal to or higher than the reference voltage Vref.

Therefore, the comparison signal PW is made logical H the first time the zero-crossing time occurs. As it is in 40 Therefore, the change timing of the detection signal Qs and the zero-crossing timing can be made a one-to-one correspondence with each other by performing the following action. The detection signal Qs is inverted when the combined signal PS coincides with the expected signal for the first time. When the switching element SW2 is turned off, the W-phase terminal voltage Vw may alternately take a value higher than and a value lower than the reference voltage Vref. Also in this case, a time when the combined signal PS and the expectation signal coincide with each other for the first time may be taken as the zero-crossing time. This is because in this case only the following takes place. The combined signal of 4 in the figure is replaced by 5.

Meanwhile, when a combined signal of one bit is generated from the comparison signals PU, PV, PW, as shown in FIG 12 is shown, the combined signal is often inverted at PWM control. Therefore, the zero crossing timing can not be detected.

In fact, the comparison signal PW can be instantaneously brought to logical H before the zero-crossing time occurs, since ringing noise in conjunction with a change in the operation of the switching elements SW1 to SW6 occurs. In this case, however, it is likely that the logical value of the comparison signal PU is different from that in FIG 43 is shown. Therefore, the possibility that the combined signal PS of three bits and the expectation signal coincide with each other before the zero-crossing time occurs is minimized.

Around reliable erroneously detecting a zero-crossing timing due to to avoid the influence of a ringing noise, it is desirable the following measure perform. Of the values of the combined signal PS, those whose Duration equal or shorter is a predetermined value, not compared with the expectation signal. This processing can be performed, for example, by performing the following measure carried out become. The combined signal PS comes with a fast sampling time sampled and values that are two or more in adjacent Scanning periods are affected by noise considered and excluded. The previous erroneous detection can just as reliable by slight Offsetting correcting the reference voltage Vref avoided generated based on virtual zero becomes.

• (1) Das kombinierte Signal (Erwartungssignal Se) der Vergleichssignale PU, PV, PW, das angenommen wird, wenn die Nulldurchgangszeit in dem vorliegenden Betriebszustand der Schaltelemente SW1 bis SW6 auftritt, wird mit dem tatsächlichen kombinierten Signal PS bezüglich jeder Phase verglichen. Auf der Grundlage der Ergebnisse von diesen Vergleichen wird eine Information erfasst, die den elektrischen Winkel des bürstenlosen Motors 2 betrifft. Daher kann verglichen mit Fällen, in denen ein kombiniertes Signal von einem Bit der Vergleichssignale PU, PV, PW verwendet wird, eine aufwendigere Information verwendet werden. Aus diesem Grund kann eine hochgenaue Information bezüglich eines elektrischen Winkels erfasst werden.
• (2) Ein Nulldurchgangszeitpunkt wird auf der Grundlage eines Übereinstimmens zwischen den angenommenen Werten der Vergleichssignale PU, PV, PW und den tatsächlichen Werten bezüglich allen der Phasen erfasst. Anders ausgedrückt wird ein Nulldurchgangszeitpunkt auf der Grundlage eines Übereinstimmens zwischen dem kombinierten Signal PS von drei Bit und dem Erwartungssignal bezüglich allen der Bits erfasst. Daher können Bedingungen zum Bestimmen eines Nulldurchgangszeitpunkts verglichen mit Fällen, in denen eine Zeit, zu der ein kombiniertes Signal von ein Bit, das die Ergebnisse eines Vergleichs bezüglich allen der Phasen anzeigt, sich ändert, als die Nulldurchgangszeit genommen wird, strenger gemacht werden. Aus diesem Grund kann ein Nulldurchgangszeitpunkt genau erfasst werden.
• (3) Ein bestimmter Zeitpunkt, der eine Grundlage zum Ändern des Betriebszustands der Schaltelemente SW1 bis SW6 vorsieht, wird auf der Grundlage eines Nulldurchgangszeitpunkts festgelegt. Daher kann ein bestimmter Zeitpunkt zweckmäßig festgelegt werden.
• (4) Ein bestimmter Zeitpunkt und ein Nulldurchgangszeitpunkt werden in eine Eins-zu-eins-Entsprechung zueinander gebracht. Als Ergebnis wird der Betriebszustand der Schaltelemente SW1 bis SW6 ebenso in eine Eins-zu-eins-Entsprechung zu dem Nulldurchgangszeitpunkt gebracht. Daher können die Vergleichssignale PU, PV, PW (Erwartungssignal), die angenommen werden, wenn die Nulldurchgangszeit in dem vorliegenden Betriebszustand der Schaltelemente SW1 bis SW6 auftritt, eindeutig bestimmt werden.
• (5) Die Referenzspannung Vref wird durch die virtuelle Nullpunktspannung des bürstenlosen Motors 2 festgelegt. Wenn ein Strom, der dem bürstenlosen Motor 2 zugeführt wird, übermäßig groß ist, wird ein PWM-Steuern in den jeweiligen Dauern eines Zulassens eines EIN-Betriebs für die Schaltelemente SW2, SW4, SW6 ausgeführt, die durch den bestimmten Zeitpunkt bestimmt werden, und wird ein Betrieb zwischen einem EIN-Betrieb und einem AUS-Betrieb geschaltet. In diesem Fall kann ein Nulldurchgangszeitpunkt nicht durch ein logisch kombiniertes Signal von einem Bit der Vergleichssignale PU, PV, PW erfasst werden. Gemäß diesem Ausführungsbeispiel der vorliegenden Erfindung kann unterdessen ein Nulldurchgangszeitpunkt auf der Grundlage eines Vergleichs des Erwartungssignals mit dem kombinierten Signal PS, von denen beide Drei-Bit-Signale sind, genau bestimmt werden.

According to this embodiment of the present invention described in detail, the following advantages can be provided.

• (1) The combined signal (expectation signal Se) of the comparison signals PU, PV, PW assumed when the zero-crossing time occurs in the present operating state of the switching elements SW1 to SW6 is compared with the actual combined signal PS with respect to each phase. On the basis of the results of these comparisons, information is detected which is the electrical angle of the brushless motor 2 concerns. Therefore, more complex information can be used compared with cases where a combined signal of one bit of the comparison signals PU, PV, PW is used. For this reason, highly accurate information regarding an electrical angle can be detected.
• (2) A zero-crossing timing is detected based on coincidence between the assumed values of the comparison signals PU, PV, PW and the actual values with respect to all the phases. In other words, a zero-crossing timing is detected on the basis of a match between the combined signal PS of three bits and the expectation signal with respect to all of the bits. Therefore, conditions for determining a zero-crossing timing can be made more stringent compared with cases where a time when a combined signal of one bit that indicates the results of comparison with respect to all the phases changes as the zero-crossing time is taken. For this reason, a zero-crossing timing can be detected accurately.
• (3) A certain timing providing a basis for changing the operating state of the switching elements SW1 to SW6 is set on the basis of a zero-crossing timing. Therefore, a specific time can be set appropriately.
• (4) A certain time and a zero-crossing time are brought into a one-to-one correspondence with each other. As a result, the operating state of the switching elements SW1 to SW6 is also brought into a one-to-one correspondence with the zero-crossing timing. Therefore, the comparison signals PU, PV, PW (expectation signal) adopted when the zero-crossing time occurs in the present operating state of the switching elements SW1 to SW6 can be uniquely determined.
• (5) The reference voltage Vref is determined by the virtual zero point voltage of the brushless motor 2 established. If a current, the brushless motor 2 is supplied, is excessively large, PWM control is executed in the respective periods of permitting an ON operation for the switching elements SW2, SW4, SW6 determined by the specific time, and becomes an operation between an ON operation and switched to OFF mode. In this case, a zero-crossing timing can not be detected by a logically combined signal of one bit of the comparison signals PU, PV, PW. Meanwhile, according to this embodiment of the present invention, a zero-crossing timing can be accurately determined on the basis of a comparison of the expectation signal with the combined signal PS, both of which are three-bit signals.

In the eleventh embodiment The present invention provides a zero crossing time on the Basis of a match between the combined signal PS and the expectation signal Se with respect to all of the bits detected. However, for example, a zero crossing time based on a match between bits corresponding to a phase in which the induced voltage and the reference voltage Vref in the PWM control each other have a zero crossing.

Twelfth embodiment

The twelfth embodiment of the present invention is similar to the ninth embodiment Example of the present invention ( 31 to 36 ).

When the battery 214 and the inverter 12 inadequately connected, resulting in an electrical separation between them, or any other similar event occurs, the following phenomenon may take place. Due to transmitting a vibration of the vehicle to the battery 214 or any other similar reason, the battery can 214 and the inverter 12 are separated from each other immediately and then routing between them is formed again. When a power supply to the brushless motor 2 is temporarily interrupted at this time, the speed of the brushless motor 2 reduced. When fuel discharged from a fuel tank to the upstream side from a fuel pump flows back at this time, a reverse rotation side force is applied to the brushless motor 2 exercised and this can eventually cause a reverse rotation. In this situation, when the switching elements SW1 to SW6 are operated as under normal conditions, an oscillation phenomenon occurs that the brushless motor 2 a normal rotation and a reverse rotation repeated. It is difficult to use the brushless motor 2 to control in a convenient rotational state.

That the brushless motor 2 is reversely rotated, can be appropriately detected on the basis of the previous combined signal PS, which consists of three bits. More specifically, as it should in 44 is shown when the brushless motor 2 is normal (in the forward direction), time series data relating to the combined signal PS coincide with time sequence data with respect to the expectation signal Se. If the brushless motor 2 Meanwhile, in the meantime, time sequential data regarding the combined signal PS should coincide with data obtained by time-reversed time series data regarding the expectation signal, as shown in FIG 45 is shown. For this reason, a reverse rotation of the brushless motor 2 be detected on the basis of the combined signal PS.

It is possible that when the rotational state of the brushless motor 2 becomes abnormal, all of the switching elements SW1 to SW6 are turned off and the operation waits until the brushless motor 2 stops. Then the brushless motor 2 restarted. In this case, however, it takes a long time to get the brushless motor 2 to return to a normal state.

To cope with this, the following processing is carried out in this embodiment of the present invention. If it is detected that the brushless motor 2 Turning backwards, processing is performed to stop the rotation of the brushless motor 2 forcibly stopping, and then a restart processing is executed. 46 Provides processing to restart the brushless motor 2 in this embodiment of the present invention. This processing is performed by the drive control circuit 220 repeatedly executed, for example, in a predetermined cycle.

This series of processing is performed as follows. In S360, it is checked if the value Cmk of the masking duration counter is zero or not. If it is determined that the value of the masking duration counter is zero, it is determined in S362 whether or not the combined signal PS of the comparison signals PU, PV, PW has changed. This processing is for determining whether or not it is the zero-crossing time. If it is determined that the combined signal PS has changed, it is determined in S364 whether or not the present combined signal coincides with the previous expectation signal before the last one. This processing is used to determine if the brushless motor 2 turns backwards or not. More specifically, as it is in 44 and 45 is shown when the brushless motor 2 reverse rotation, time sequence data regarding the combined signal PS reversed. Therefore, it is assumed that the present combined signal PS coincides with the expectation signal before the last one. If the present combined signal PS coincides with the expectation signal Se before the last one, it is determined in S366 that the brushless motor 2 turns backwards.

In S268, processing is subsequently performed to control the rotation of the brushless motor 2 forcibly stop. More specifically, the switching elements SW1, SW3, SW5 or the switching elements SW2, SW4, SW6 are all turned on to all of the phases of the brushless motor 2 to close briefly. Therefore, a current through the brushless motor 2 supplied only by an induced voltage, which in conjunction with the rotation of the brushless motor 2 is produced. This current is rapidly attenuated by the resistance value of the current passage and the like. As a result, the rotational energy of the brushless motor 2 converted to electrical energy and then attenuated. For this reason, the brushless motor 2 be stopped quickly.

When the speed of the brushless motor 2 is substantially zero (S370: YES), in S372 becomes Restart processing performed. The speed of the brushless motor 2 is calculated based on time intervals between adjacent zero crossing times. This can be done using the maximum value of Cm of the meter.

If in any of S360 to S364, a negative determination is made or when the processing of S372 is finished, this is the row finished processing once.

According to this embodiment of the present invention the following advantages in addition to the advantages (1) to (5) of the eleventh embodiment of the present invention Invention be provided.

• (6) The rotation state of the brushless motor 2 is determined to be abnormal based on a disagreement between the combined signal PS assumed when the zero-crossing time occurs in the present operating state of the switching elements SW1 to SW6 and the expectation signal. The use of the combined signal PS of three bits and the expectation signal Se of three bits makes it possible to appropriately determine the presence or absence of an anomaly.
• (7) The presence of an anomaly that is the brushless motor 2 conversely, it is determined on the basis of a match therebetween, which is achieved by temporally reversing the time series pattern of the combined signal PS assumed by the time series pattern of the operating state of the switching elements SW1 to SW6 and the actual time sequence pattern. Therefore, it can be appropriately detected that the brushless motor 2 reversed turns.
• (8) If it is determined that the brushless motor 2 Turning in reverse, processing is performed to the brushless motor 2 forcibly stop, and then becomes the brushless motor 2 restarted. Therefore, the brushless motor 2 be quickly returned to a normal state.
• (9) Conduction will be of all the phases of the brushless motor 2 to either the positive pole or the negative pole of the battery 214 formed to forcibly the brushless motor 2 to stop. Therefore, the rotational speed of the brushless motor 2 be reduced quickly.

In the twelfth embodiment of the present invention, all of the phases of the brushless motor become 2 shorted to the brushless motor 2 forcibly stop. Instead, switching of the switching elements SW1 to SW6 may be controlled to generate a torque for stopping the rotation. Further, the reverse rotation is detected under the condition that the combined signal PS coincides with the expectation signal before the last one. A reverse rotation may be detected in addition to this condition if the next combined signal PS coincides with the expectation signal preceding the expectation signal before the last one.

Thirteenth embodiment

A thirteenth embodiment of the present invention is directed to improving the following problem. That is, if any of phase lines of the brushless motor 2 is interrupted, puts the inverter 12 a voltage to the lines of the motor 2 which are not interrupted. However, it is likely that a flow of the current will be affected and an excessive load on the brushless motor 2 is exercised. It is therefore of the JP 2-290191A proposed to determine, using a short-circuit resistance, whether a phase current is flowing and to detect a presence / absence of an interrupt based on this determination. In this case, however, a detection element is required to detect a voltage drop across the short-circuit resistor. This becomes the size of the circuit of the drive control circuit 220 increase.

Therefore, according to this embodiment of the present invention, the presence / absence of disconnection is detected on the basis of the comparison signals PU, PV, PW. Such comparison signals PU, PV, PW are from the control circuit 220 to be used to operate the switching elements SW1 to SW6. Therefore, by detecting the break based on such signals, an increase in the size of the circuit can be avoided. The basis of detecting the interruption based on the comparison signals PU, PV, PW will be described first.

As previously described, the comparison signals PU, PV, PW change as in (b) of FIG 43 is shown. Here, the reference voltage Vref exceeds the voltage VB of the positive pole of the battery 214 When the switching elements, which are PWM controlled, switched from ON to OFF who the. As a result, the comparison signal of a phase in which the switching element continues to be ON is reversed in its logical value. If the brushless motor 2 is in the interrupted state, as it is in 47 is shown, the comparison signal is not reversed in its logical value. Here, (a) and (b) correspond to 47 (a) and (b) from 43 and illustrate a case in which the phase line of a W phase of the brushless motor 2 interrupted on one side, which is closer to the brushless motor 2 as a connection point with the resistive element RW.

As it is in 47 is shown, the terminal voltage Vw of the W phase drops to approximately the voltage of the negative pole of the battery 214 , This is because, although the W phase is in the high impedance state in the example that is in FIG 47 is shown, the potential of the W phase to the negative potential of the battery 214 due to a parasitic capacitance between the gate and the drain of the switching element SW6 and the like is reduced. In this case, when the switching element SW2 is turned from off to on, a current flows in the diode D1. As a result, the terminal voltage Vu of the U-phase increases above the voltage VB of the positive pole of the battery 214 However, the reference voltage Vref falls below the voltage VB of the positive pole of the battery 214 , This is because the terminal voltage of the W phase is reduced by the interruption.

For this reason, regardless of the state of the switching element SW2, the reference voltage Vref continues to be lower than the terminal voltage Vu of the V-phase connected to one side of the positive-pole voltage VB of the battery 214 connected is. Therefore, as in (b) of 47 4, the V-phase comparison signal PV continues to be H, and is different from the state shown in (b) of FIG 43 is shown. Therefore, the break of the brushless motor 2 be detected based on the difference between these states.

Although 47 FIG. 10 illustrates a case where a break occurs in a phase (to be changed to the high impedance) in which the switching elements in both the high-side branch and the low-side branch are turned off, the interruption may be made to the like Way can be detected using the comparison signal PV even if a phase (U-phase in 47 ) which is PWM controlled. This is because the reference voltage Vref is the voltage VB of the positive pole of the battery 214 does not exceed, since the terminal voltage Vu of the U-phase of the battery 214 continues, the voltage of the negative pole of the battery 214 to be.

48 represents a processing of detecting an interrupt. This processing is repeated by the control circuit 220 for example, executed in a predetermined duration.

In a series of this processing, it is checked in S380 whether the PWM control performed in S352 in FIG 42 is executed is executed. When it is determined that the PWM control is being executed, a phase which is composed of the switching elements SW1, SW3, SW5 in the high side branch in the period of permitting ON operation is determined in S382. Through this step, the V phase in the example of 47 certainly. In a subsequent S384, it is checked whether or not it is the duration of permitting an ON operation of the specific phase. This step is for determining a duration in which the terminal voltage of the appropriate phase to be used for detecting an interruption does not change.

If it is determined in S384 that it is in the period of permitting an ON operation, it is further checked in S386 whether the logical value of the phase comparison signal determined in S382 is L large. This step is for determining the presence / absence of an interrupt. If it is affirmatively determined in step S386, an L-detection flag is set to ON in S388 to indicate that the logical value has become L. This step may be as a step of changing a register value in the control circuit 220 be executed.

If S388 has been completed and a negative determination has been made in S386, the processing returns to S390. In S390, it is checked whether the L detection flag is on or not. This step is to determine the presence / absence of the interruption. That is, if the logical value of the comparison signal does not become L within the period of permitting one of the phases determined in S382, it is considered that the phenomenon occurred by (b) of 47 is shown. In this case, it is determined that the interruption has occurred. Therefore, if a negative determination is made in S390, a notification of detection of an interruption of control devices to an external side is issued in a step S392. If S392 has ended or a negative determination is made in step S380 or S390 has been set, the L-detection flag is set to OFF in S394.

If S394 has been completed, the series of processing is finished.

• (10) Unter der Bedingung, dass lediglich eine Phase des bürstenlosen Motors 2 zu dem Anschluss des negativen Pols der Batterie 214 leitend gemacht wird und eine andere Phase zu dem Anschluss des positiven Pols der Batterie 214 leitend gemacht wird, wird ein Vorhandensein/Nichtvorhandensein der Unterbrechung des bürstenlosen Motors 2 auf der Grundlage des Vorhandenseins/Nichtvorhandenseins einer Umkehr des Vergleichssignals der anderen Phase zu der Zeit eines Ausschaltens des Schaltelements erfasst, was den Anschluss des negativen Pols und den bürstenlosen Motor 2 leitend macht. Daher kann eine Unterbrechung des bürstenlosen Motors 2 erfasst werden.
• (11) In dem 120°-Erregungssteuern überschreitet die Referenzspannung Vref nicht die Spannung des positiven Pols der Batterie 214. In diesem Fall wird, da die Unterbrechung nicht auf der Grundlage des Vergleichssignals erfasst werden kann, die Unterbrechung in dem PWM-Steuern erfasst. Deshalb kann ein Erfassen einer Unterbrechung zweckmäßig erzielt werden.

According to this embodiment of the present invention, the following advantages are achieved in addition to the advantages of the eleventh embodiment of the present invention.

• (10) Under the condition that only one phase of the brushless motor 2 to the connection of the negative pole of the battery 214 is made conductive and another phase to the connection of the positive pole of the battery 214 is rendered a presence / absence of the break of the brushless motor 2 detected on the basis of the presence / absence of a reversal of the comparison signal of the other phase at the time of turning off the switching element, which means the connection of the negative pole and the brushless motor 2 makes conductive. Therefore, a break in the brushless motor 2 be recorded.
• (11) In the 120 ° excitation control, the reference voltage Vref does not exceed the voltage of the positive pole of the battery 214 , In this case, since the interruption can not be detected on the basis of the comparison signal, the interruption in the PWM control is detected. Therefore, detection of an interruption can be appropriately achieved.

Fourteenth embodiment

The fourteenth embodiment of the present invention is shown in FIG 49 shown. The brushless motor 2 is a two-phase motor. For detecting presence / absence of an interruption in the two-phase motor 2 In the similar manner as in the thirteenth embodiment of the present invention, a series circuit of diodes D5, D6 is connected in parallel with a series circuit of the switching elements SW1, SW2 and a series circuit of the switching elements SW3, SW4. A connection point between the diodes D5, D6 is at the zero point of the brushless motor 2a connected. Therefore, a phase connected to a connection point between the switching elements SW1, SW2 is defined as a U phase, a phase connected to a connection point between the switching elements SW3, SW4 is defined as a V phase and is a phase connected to the connection point between the diodes D5, D6 is defined as a W phase.

One Comparator Cu generates a comparison signal PU by comparison the terminal voltage Vu and the reference voltage Vref. A comparator Cv generates a comparison signal PV by comparing the terminal voltage Vv and the reference voltage Vref. A comparator Cw generates one Comparison signal PW by comparing the terminal voltage Vw with the reference voltage Vref. On the basis of these comparison signals PU, PV, PW can interrupt in the similar way as in the thirteenth embodiment of the present invention.

The eleventh to fourteenth embodiments of the present invention as described below.

Of the certain time is set by adjusting the decrementing speed of the meter to set a particular time relative to the rate of increment of the measuring counter established. However, you can these counters in the counting speed can be made identical and the initial value of the counter to Defining a specific time according to the value relative to the measuring counter before initialization (maximum value). If the certain time is set at a time around 30 ° from delayed the zero crossing time For example, 1/2 of the maximum value of the meter may be as the initial value of the counter to set a specific time.

The Masking time is set by adjusting the decrementing speed of the masking duration counter fixed relative to the incrementing speed of the measuring counter. However, you can these counters in the counting speed are made identical and the initial value with respect to the Masking period counter according to the value of the measuring counter set before initialization (maximum value). If an angular range from a zero crossing time to 45 ° as the masking duration can be 3/4 of the maximum value of the measuring counter as the initial value with respect to the Masking period counter be taken.

The anomaly of the rotating state of the brushless motor 2 is not limited to reverse rotation. It is only important that if the combined signal PS does not match the expectation signal coincides, the rotational state is determined as abnormal.

The PWM control is executed when the phase current of the brushless motor 2 exceeds the threshold. However, it is also possible, for example, to forcibly turn off the switching element of the low side branch, which is ON in the period of ON, only when the current continues to exceed the threshold value. An operating means for repeating switching on and off of the switching element in the period of permitting ON may therefore be provided by this control.

The Operating equipment is not limited to equipment which operates the switching elements of the branch of the low side, but can be a device that controls the switching elements of the branch high side operates. In this case, in the thirteenth and fourteenth embodiments of the present invention, the presence / absence a break based on the presence / absence a reversal of the comparison signal of the phase detected, in which the switching element of the branch of the low side to the ON state is fixed.

The reference voltage Vref need not be the virtual zero formed on the basis of the terminal voltages Vu, Vv, Vw, but may be the zero-point voltage of the brushless motor 2 be. Also in this case, the same advantages as those according to the eleventh embodiment of the present invention can be provided. Even if a 1/2 of the voltage of the battery 214 is used for the reference voltage Vref, the following can be realized by using the combined signal of three bits and the expectation signal of three bits. An anomaly of the rotation state can be accurately detected and the accuracy of detecting the zero-crossing timing can be improved in a 120 ° excitation control. If a phase line closer to one side of the inverter 12 as at one side of the connection point with the resistive elements RU, RV, RW of the phase lines of the brushless motor 2 is an interrupt can be detected using the reference voltage Vref as the zero voltage based on the same phenomenon as in the thirteenth embodiment of the present invention.

The switching elements SW1, SW3, SW5 on the high side of the respective branches may be constructed of an N-channel MOS transistor. The power supply, with the brushless motor 2 connected, no battery needs 214 but can be a generator. The brushless motor 2 does not have to be an actuator of an on-board fuel pump, but may be an actuator of an in-vehicle cooling fan.

The multi-phase lathe need not be a three-phase brushless motor and may be an engine of any number of phases. Furthermore, it does not have to be an engine, but can be a generator. Even if the number of phases of the lathe in the thirteenth embodiment of the present invention is changed to be large N (> 3), an interruption in the PWM control can be detected in the same manner as in the thirteenth embodiment of the present invention, as long as the switching element is of only one phase in the branch of a low side in the period of a permit of ON. That is, when the switching element of only one phase is changed from ON to OFF, the reference voltage Vref becomes approximately (N-1) × VB / (N + Vf). Unless the phase number N becomes excessively large, the reference voltage Vref remains lower than the voltage VB of the positive pole of the battery 214 , On the other hand, when no interruption occurs, the reference voltage Vref becomes higher than the voltage VB of the positive pole of the battery 214 , As a result, the interruption can be detected on the basis of the presence / absence of a reversal of the logical value of the comparison signal of a phase corresponding to a switching element of the high side branch set to the ON state.

Fifteenth embodiment

The fifteenth embodiment of the present invention, which is shown in FIG 51 is designed to be a brushless motor 2 , an inverter 12 and a drive control circuit 220 exhibit. This structure is similar to the eleventh embodiment of the present invention, which is shown in FIG 38 is shown, and therefore no detailed description is given.

52 FIG. 12 illustrates the manner in which the shift control of the shift control device 227 in which 120 ° excitation control is executed. More specifically, (a) represents the transition of the U-phase terminal voltage Vu shown by a solid line, the transition of the induced voltage of the U-phase represented by a two-dot chain line, and the reference voltage Vref is represented by a single-dotted line. (b) illustrates the transition of a signal (zero-crossing detection signal) Un, which relates to detecting a zero-crossing timing. (c) represents the transition of an activation signal for the switching element SW1, and (d) represents the transition of an activation signal for the switching element SW2. The modes for the V-phase and W-phase switching control are the same as those for the U Phase switching control and the explanation and description thereof are omitted.

As it is in 2 is shown, the terminal voltage Vu is correct with the positive pole or the negative pole of the battery 214 match when the switching element SW1, SW2 is ON. Meanwhile, in periods (duration of release of an induced voltage) during which both the switching element SW1 and the switching element SW2 are OFF, there is a period during which a current is not passed through the U-phase. In these periods, the terminal voltage Vu is equal to the induced voltage. Even in the periods during which both the switching element SW1 and the switching element SW2 are OFF, the terminal voltage Vu does not coincide with the induced voltage when a current is supplied through the diodes D1, D2 (commutation transition state).

Out For this reason, the time at which the terminal voltage Vu and the reference voltage Vref agree with each other in the following period, the zero crossing time at which the induced voltage Vu and the reference voltage Vref agree with each other: a duration, during which both the switching element SW1 and the switching element SW2 OFF are and no electricity over the diodes D1 or D2 is passed. That's why the time from the zero crossing instant by a predetermined electrical Angle (for example 30 °) delayed is defined as the specific time, and that particular time Time is taken as a time at which the operation of the switching elements SW1, SW2 from an OFF operation to an ON operation changed becomes. The ON state becomes for one Duration of 120 ° of an occurrence of the specific time continued. More precisely The following specific time will be the time to turn on of the switching element SW1 of the high-side branch: a certain time by a predetermined electrical angle (for example 30 °) is delayed from the zero crossing time to which the induced Voltage coincides with the reference voltage Vref in their rise process. The following specific time is considered the time to turn on of the switching element SW2 of the low-side branch uses: the specific time, by a predetermined electrical Angle (for example 30 °) is delayed from the zero crossing time to which the induced Voltage coincides with the reference voltage Vref in their waste process. The timing for turning on the U-phase switching elements SW1, SW2 may be determined by the zero-crossing timing in the rising process of the induced voltage of the U phase and the zero crossing time in their waste process. In this embodiment For this reason, the present invention becomes the zero-crossing detection signal Un rising at the zero-crossing timing in the rising operation and drops to the zero-crossing timing in the fall-off process. Its leading edges and trailing edges are used to create a certain time.

53 FIG. 12 illustrates the manner in which the shift control is performed in a normal state in which the rotational speed of the brushless motor 2 is stabilized. More specifically, (a) represents the transition of the terminal voltage Vu, Vv, Vw; represents (b) the transition of comparison signals Uc, Vc, Wc representing the amplitude relationship between the terminal voltages and the reference voltage Vref; (c) represents the transition of the zero-crossing detection signals Un, Vn, Wn; represents (d) the transition of the values with respect to different counters; and (e) represents the transition of the activation signals for the switching elements SW1 to SW6. The activation signals shown in (e) include activation signals U +, V +, W + for the switching elements SW1, SW3, SW5 of the high side branches and activation signals U, V, W- for the switching elements SW2, SW4, SW6 of the branches of the low side. The switching elements SW1, SW3, SW5 of the high-side branches are P-channel transistors; therefore, the durations for which the activation signals U +, V +, W + are at logic L are the durations for which they are ON.

The Solid lines in (d) show the value Cm of a measurement counter for Measuring a time interval between adjacent appearances of a Zero crossing time. As shown, the meter reading becomes initialized at any time, at which time the zero crossing point occurs and restarts a time counting operation. A time interval between adjacent occurrences of a zero crossing time point has a correlation to a speed. That's why the value of the counter, immediately before being initialized (the maximum value of the counter) one Parameter, which has a correlation to a rotational speed.

The single-dotted lines in (d) show the value Cs of one counter to set a specific time that counts a time that from then on, when the zero crossing time occurs, until then is required when the specific time occurs and lays thereby determining a certain time. The counter for specifying a specific Time takes the value of the counter before initializing as its initial value on occurrence a zero crossing time and decrements it. Then he sets the time at which the value is zero, as a specific Time fixed. At this time, the following operation is performed. If the interval between the zero crossing time and the determined one Time is for example 30 °, the decrementing speed becomes twice the incrementing speed of the measuring counter established. This is so because there are intervals between the adjacent ones occurring events of the zero crossing time point are 60 °. For this reason, if the speed is constant, the time to which the value of the counter to set a given time becomes zero, equal to one Be at 30 ° from delayed the zero crossing time is.

The double-dashed lines in (d) show the value Cmk of one Masking period counter. The masking duration counter determines a duration (masking duration) for which detection of the zero-crossing time point based on the comparison of the terminal voltages Vu, Vv, Vw is locked to the reference voltage Vref with respect to an amplitude (not released). This counter is used to prevent the following event. If the terminal voltages Vu, Vv, Vw during a Duration, for which is a current over the diodes D1 to D6 supplied is equal to the reference voltage Vref becomes an occurrence of a zero crossing time detected incorrectly. This counter takes also the value of the counter initializing as an initial value upon occurrence of the Zero crossing time and decrements it. Then lay down It sets the duration before the value is made zero as a masking period. If the masking period is for a duration of one occurrence of a Zero crossing time is set to, for example, 45 °, the decrementing speed set to 3/2 times the incrementing speed of the measuring counter.

The processing of a switching control will be described with reference to FIG 54 and 55 described. 54 represents a processing for setting the counter values with respect to the previous three counters. This processing is performed by drive control circuit 220 repeatedly executed, for example, in a predetermined cycle.

These Series of processing is carried out as follows. In S410 it is checked if the Value Cmk of the masking duration counter Zero or not. If it is determined that the value is zero, it is determined in S412 whether any of the comparison signals Uc, Vc, Wc has been inverted or not. In fact, this processing serves for determining whether the previous zero-crossing detection signal Un, Vn, Wn is at its leading edge or its trailing edge. If it is determined in S412 that any signal is inverted is the value Cs of the counter as the values Cs and Cmk of the counter for setting a specific time and the mask duration counter in S416 taken. In S416 the measuring counter is initialized (Cm = 0).

If in S410 or S412 a negative determination is made, becomes the meter incremented in S418. In S420 it is checked below if the Cs value of the counter to set a specific time is zero or not. If the value of the counter To set a specific time is not zero, the counter decremented to set a particular time in S424. Meanwhile, it is when affirmative determination is made in S420 or when the processing of S422 is finished, determined in S424, whether the value of the masking duration counter is zero or not. If the value of the masking duration counter is not zero, then the masking duration counter decremented in S426.

If in S424 an affirmative determination is made or if the processing finished by S416 or S426, this series is a processing ended once.

5 FIG. 14 illustrates a flow of processing for turning on the switching elements SW1 to SW6. This processing is performed by the drive control circuit 220 for example, repeatedly executed in a predetermined cycle.

The series of a processing in 55 is performed as follows. In S430, it is checked whether the value Cs of the counter for setting a specific time is zero or not. This processing is for determining whether the specific time has occurred or not. When it is determined that the value of the counter for setting a specific time has been made zero, the processing is executed to turn on any one of the switching elements SW1 to SW6. Exactly ge when the zero crossing timing at which the value Cm of the counter is set as the value of the counter for setting a specific time is the U-phase zero-crossing timing (S432: YES), the following processing is executed. When the zero-crossing detection signal rises (S434: YES), the switching element SW1 is turned on (S436). When the zero-crossing detection signal drops (S434: NO), the switching element SW2 is turned on (S438).

meanwhile when the previous zero crossing time is a V-phase zero-crossing time is (S440: YES), the following processing is executed. When the zero-crossing detection signal rises (S442: YES), the switching element SW3 is turned on (S444). If the zero crossing detection signal drops (S442: NO), the switching element SW4 is turned on (S446). If the previous zero crossing time is a W phase zero crossing time is (S440: NO), the following processing is executed. If the zero-crossing detection signal rises (S448: YES), it becomes Switching element SW5 switched on (S450). When the zero crossing detection signal drops (S448: NO), the switching element SW6 is turned on (S452).

The method for turning off the switching elements SW1 to SW6 is similar to the previous one. That is, they are turned off at a time delayed by a predetermined electrical angle (for example, 30 °) from a specific specific time, as shown in FIG 53 is shown. More specifically, the switching elements SW1, SW3, SW5 of the high side branches are turned off at the time delayed by a predetermined electrical angle from the zero crossing timing in the rising phase of a different phase induced voltage. The switching elements SW2, SW4, SW6 of the low-side branches are turned off at the time that is delayed by a predetermined electrical angle from the zero-crossing timing to the falling-off operation. This processing may be similar to the processing in 55 and the description of it will be omitted.

As it is in the 56A and 56B As an example of acceleration, the zero crossing time depends on the speed of the brushless motor 2 from. Because of this, a problem arises when the speed of the brushless motor 2 fluctuates. The rotational speed in a time interval between an occurrence of a zero-crossing timing at the time of initialization and the next occurrence of the next zero-crossing timing can not be accurately represented by the value of the counter immediately before initialization. For this reason, there is a possibility that the timing at which the value Cs of the counter for setting a certain timing is made zero deviates from the specific timing.

As an example of a case where a rotational speed fluctuates, the 57A and 57B is the result of an experiment relating to a timing error of the particular time point at an acceleration. When the initial speed has been increased from zero to a predetermined speed as shown in FIG 57A is shown, a detection error (a phase shift) of the angle of the rotor has been observed as shown in FIG 57B is shown. It goes without saying 57B in that the phase shift is particularly large in the region of low speed.

In this embodiment of the present invention, therefore, information that is a change in the rotational speed of the brushless motor 2 is extracted from the result of detecting the zero-crossing timing, and the specific timing is set on the basis of the extracted information. More specifically, a change in the rotational speed included in the preceding information is associated with a duration before occurrence of the specific timing. However, it is assumed that the change of the rotational speed may be used to determine the difference between a rotational speed in the time interval between adjacent occurrence events of a zero-crossing timing immediately before occurrence of the determined timing and a rotational speed immediately before the occurrence of the specific timing. Consequently, the previous difference is determined on the basis of the previous information, and a specific time is set.

58A Fig. 15 illustrates processing for setting a specific time in this embodiment of the present invention. This processing is performed by the drive control circuit 220 for example, repeatedly executed in a predetermined cycle.

This series of processing is performed as follows. In S460, it is checked if it is the time to cancel a masking or not. That is, it is checked whether the value CmK of the masking duration counter has been made zero or not. If it is the time to erase a mask, For example, the value Cac of an acceleration detection counter is incremented in S462. In S464, it is checked whether any of the comparison signals Uc, Vc, Wc as in S412 in FIG 54 has been inverted or not. If an affirmative determination is made in S464, it is determined in S466 whether or not the value Cac of the acceleration detection counter is equal to or greater than a predetermined value B. This processing is to determine if the brushless motor 2 is in an acceleration state or not. In this example, when the masking period is a duration from an occurrence of a zero-crossing timing to 45 ° and the incrementing speed of the acceleration detection counter is identical to the incrementing speed of the measurement counter, the following can be realized. If the value of the acceleration detection counter is less than 1/4 times the previous counter, it can be determined that the brushless motor 2 is in an accelerated state. In this case, for this reason, the predetermined value B is set to a value smaller than 1/4 times the previous value of the counter.

If it is determined in S466 that the brushless motor 2 is in an acceleration state, a correction value .DELTA.A.sub.i for correcting the value of the counter for setting a specific time in accordance with the value of the acceleration detection counter is set in S468. Acceleration is increased with decreasing the value of the acceleration detection counter. In consideration of this, in this example, the correction value ΔAi is set to a larger value as the value Cac of the acceleration detection counter is decreased as shown in FIG 58B is shown. In S470, subsequently, a value obtained by subtracting the correction value ΔAi from the value Cs of the specific-time setting counter C, is set as a value Cs of the predetermined-time setting counter.

If in S466 a negative determination is made or if the processing S470 is finished, the acceleration detection counter in S472 initialized (Cac = 0). If in S460 or S464 a negative Determination performed or if the processing of S472 is finished, this is Series of processing finished once.

59 provides the way to control the output of the brushless motor 2 in this embodiment of the present invention. More specifically, (a) represents the transition of the angle of the rotor; represents (b) the transition of a speed; and (c) illustrates the transition of a phase current. It should be understood that a detection error of a given instant in acceleration is also small (for example, a duration of 0.04 to 0.06 seconds). Therefore, the speed can be easily increased to a desired speed. When the correction processing for the meter counter, which is in 58A meanwhile, the error of the particular time is large as it is in 60 is shown. Therefore, the speed can not be easily increased.

• (1) Eine Information, die eine Änderung der Drehzahl des bürstenlosen Motors 2 betrifft wird aus dem Ergebnis eines Erfassens des Nulldurchgangszeitpunkts extrahiert und ein bestimmter Zeitpunkt wird auf der Grundlage dieser Information festgelegt. Aus diesem Grund kann eine Zeit, die von einem bestimmten Auftreten eines Nulldurchgangszeitpunkts zu einem Auftreten eines bestimmten Zeitpunkts erforderlich ist (Wert Cs des Zählers zum Festlegen eines bestimmten Zeitpunkts) genau berechnet werden. Folglich kann ein bestimmter Zeitpunkt genau festgelegt werden.
• (2) Eine Zeit (ein Anfangswert des Zählers zum Festlegen eines bestimmten Zeitpunkts), die von dem Auftreten eines Nulldurchgangszeitpunkts unmittelbar vor einem Auftreten eines bestimmten Zeitpunkts zu dem Auftreten eines bestimmten Zeitpunkts erforderlich ist, wird aus dem folgenden Intervall berechnet: ein Intervall zwischen Auftrittvorgängen eines Nulldurchgangszeitpunkts (der Wert des Zählers). Nun wird sie auf der Grundlage der vorhergehenden Information korrigiert. Daher kann eine Zeit (der Anfangswert des Zählers zum Festlegen eines bestimmten Zeitpunkts), die erforderlich ist, auch dann genau berechnet werden, wenn die Drehzahl schwankt.
• (3) Die vorhergehende Information wird auf der Grundlage einer Zeit von da an, wenn eine Maskierung ausgelöscht wird, bis zu dahin, wenn ein Nulldurchgangszeitpunkt auftritt, erfasst. Daher kann die vorhergehende Information zweckmäßig erfasst werden.
• (4) Ein bestimmter Zeitpunkt eilt mehr voraus, wenn die Zeit, bis der Nulldurchgangszeitpunkt auftritt, kürzer wird. Daher kann der bestimmte Zeitpunkt gemäß einer Beschleunigung genau festgelegt werden.

According to the fifteenth embodiment of the present invention, the following advantages can be provided.

• (1) An information indicating a change in the rotational speed of the brushless motor 2 is extracted from the result of detecting the zero-crossing timing, and a specific timing is set on the basis of this information. For this reason, a time required from a specific occurrence of a zero-crossing timing to an occurrence of a certain timing (value Cs of the counter for setting a specific time) can be accurately calculated. Consequently, a specific time can be specified precisely.
• (2) A time (an initial value of the set time setting counter) required from occurrence of a zero-crossing point immediately before occurrence of a certain time point at the occurrence of a certain time point is calculated from the following interval: an interval between occurrence times a zero crossing time (the value of the counter). Now it is corrected on the basis of the previous information. Therefore, a time (the initial value of the set time setting counter) required may be accurately calculated even if the speed fluctuates.
• (3) The previous information is detected on the basis of a time from when masking is extinguished to when a zero-crossing timing occurs. Therefore, the previous information can be appropriately detected.
• (4) A certain time precedes more when the time until the zero-crossing time occurs becomes shorter. Therefore, the specific time can be set accurately according to acceleration.

Sixteenth embodiment

In the sixteenth embodiment of the present invention, the acceleration of the brushless motor 2 is calculated from a result of detecting a zero-crossing timing, and a specific timing is variably set according to this acceleration. 61A Fig. 15 illustrates processing for setting a specific time in this embodiment of the present invention. This processing is performed by the drive control circuit 220 for example, repeatedly executed in a predetermined cycle.

This series of processing is performed as follows. In S480 and S482, the processing of S410 and S412 becomes 54 to determine if it is a zero crossing time or not. If it is a zero-crossing timing, processing for calculating an acceleration Ai is performed in S484. At this time, the acceleration Ai is calculated by the following difference: a difference between the reciprocal N (i) of the time interval between the previous occurrence of the zero-crossing timing and the present occurrence of the zero-crossing timing and the reciprocal N (i-1) between the second preceding occurrence of the zero crossing time and the previous occurrence of the zero crossing time point. In fact, this processing of subtracting the reciprocal of the previous maximum value of the measurement counter from the reciprocal of the present maximum value of the measurement counter can be performed. More specifically, the acceleration ΔAi can be calculated as follows, where Ti is defined as a previous zero-crossing time. Ai = N (i) - N (i-1), N (i) = 1 / {Ti - T (i-1)}

In S486, a correction value .DELTA.A.sub.i for correcting the value Cs of the certain-time setting counter according to the calculated acceleration Ai is subsequently calculated. At this time, the correction value ΔAi is determined as shown in FIG 61B is shown. That is, when the acceleration Ai is equal to or greater than a predetermined value A2 (> 0), the correction value ΔAi is set to a negative value, so that its absolute value increases with an increase in the acceleration Ai. When the acceleration Ai is equal to or smaller than a predetermined value A1 (<0), the correction value ΔAi is set to a positive value, so that its absolute value increases with a decrease in the acceleration Ai.

In S488 will subsequently be the value Cs of the counter for specifying a particular Time by adding the correction value .DELTA.Ai to the value Cs of the counter to Setting a specific time corrected. If in S480 or S482 a negative determination is made or if the processing of S488 is finished, this series of processing is completed once.

• (5) Die Beschleunigung Ai des bürstenlosen Motors 2 wird auf der Grundlage von mehreren Werten bezüglich Zeitintervallen zwischen Auftrittvorgängen eines Nulldurchgangszeitpunkts berechnet und ein bestimmter Zeitpunkt wird auf der Grundlage dieser Beschleunigung festgelegt. Daher kann ein bestimmter Zeitpunkt genau unberücksichtigt einer Schwankung einer Drehzahl festgelegt werden.

According to the sixteenth embodiment of the present invention, the following advantage can be provided in addition to the advantages (1) and (2) of the fifteenth embodiment of the present invention.

• (5) The acceleration Ai of the brushless motor 2 is calculated on the basis of a plurality of values regarding time intervals between occurrences of a zero-crossing timing, and a specific timing is set based on this acceleration. Therefore, a specific time can be set precisely regardless of a fluctuation of a rotational speed.

Seventeenth embodiment

When a zero-crossing timing occurs before the masking-duration-counter value Cmk is made zero, it is impossible to appropriately set a certain timing and appropriately set the masking duration counter and the like. To cope with this, this seventeenth embodiment of the present invention is constructed such that a time Te elapsed from the occurrence of a zero-crossing timing to the present time is estimated based on an induced voltage at that time. 62 represents processing in the previous estimation. This processing is performed by the drive control circuit 220 for example, repeatedly executed in a predetermined cycle.

This series of processing is performed as follows. In S490, it is checked whether or not it is the time for canceling a masking when the value Cmk of the masking duration counter is changed to zero. If it is the time for canceling masking, it is determined in S492 whether occurrence occurs a zero crossing time has been completed or not. This processing can be performed by performing the following action. When an induced voltage is in its rising operation, it is checked whether a terminal voltage has already exceeded the reference voltage Vref or not. When the induced voltage is in its fall, it is checked whether the terminal voltage has already fallen below the reference voltage Vref or not.

When it is determined that an occurrence of a zero-crossing timing has already been completed, the processing proceeds to S494. In S494, a time Te elapsed from the occurrence of a zero-crossing timing to the present time is estimated on the basis of the previous value of the counter and the terminal voltage. When masking is canceled, the terminal voltage indicates the induced voltage. The difference between the value of the induced voltage and the reference voltage Vref contains information concerning the previous elapsed time Te. However, since the amplitude of an induced voltage depends on the rotational speed, it is impossible to accurately determine the elapsed time only by the present induced voltage. Thus, in this embodiment of the present invention, the elapsed time is estimated on the basis of the previous maximum value of the count value as a parameter having a correlation with the rotational speed and the present induced voltage. At this time, for example, the following measure can be performed. The shift control device 227 is composed of a microcomputer, and a data map defining the relationship between the previous maximum value of the counter and the present induced voltage and the elapsed time is used to estimate the elapsed time.

In S496 becomes a value obtained by subtracting the elapsed time from the present value of the counter is obtained as the values Cs and Cmk of the counter for specifying a particular Time and the masking duration counter set. This processing is used to set a specific time and a masking time based on an estimated Zero crossing timing. In S498, the value Cm of the counter becomes set the elapsed time. If in S490 or S492 an affirmative determination carried out or if the processing of S498 is finished, this is Series of processing finished once.

In the seventeenth embodiment The present invention takes a time of occurrence a zero crossing instant immediately before the present Time has elapsed, based on an induced voltage and a speed estimated. However, for example, an elapsed time may be based on the maximum value or minimum value of an induced voltage the previous duration of canceling a masking and the present induced voltage can be estimated. More precisely is a speed under consideration whose uses the amplitude of an induced voltage depends on the speed. The amplitude can also be determined using the previous maximum value or minimum value of the induced voltage instead become.

• (6) Wenn das Auftreten eines Nulldurchgangszeitpunkts unmittelbar zuvor bereits beendet worden ist, wenn ein Maskieren ausgelöscht wird, wird eine Zeit, die von dem Auftreten des Nulldurchgangszeitpunkts unmittelbar zuvor bis zu der vorliegenden Zeit verstrichen ist, auf der Grundlage einer induzierten Spannung geschätzt. Daher kann eine Zeit, die von dem Auftreten eines Nulldurchgangszeitpunkts unmittelbar zuvor bis zu einem Auftreten eines bestimmten Zeitpunkts erforderlich ist, geschätzt werden.

According to this embodiment of the present invention, the following advantages can be provided in addition to the advantages (1) to (4) of the fifteenth embodiment of the present invention.

• (6) When the occurrence of a zero-crossing timing has already been terminated immediately before when masking is canceled, a time elapsed from the occurrence of the zero-crossing timing immediately before to the present time is estimated on the basis of an induced voltage. Therefore, a time required from the occurrence of a zero-crossing timing immediately before to a occurrence of a certain timing can be estimated.

Eighteenth embodiment

In the eighteenth embodiment of the present invention, a threshold value for the currents that are passed through the switching elements SW1 to SW6, according to the acceleration of the brushless motor 2 changeable. 63 represents a processing for setting the previous current limit value. This processing is performed by the drive control circuit 220 for example, repeatedly executed in a predetermined cycle.

This series of processing is performed as follows. In S500, an acceleration Ai is calculated. This processing is the same as the processing of S484 in 61 , In S502, it is subsequently checked whether the acceleration Ai is equal to or greater than a first specific acceleration Amax or not. The first specific acceleration Amax is set in accordance with too high an acceleration, at which a deterioration of the accuracy of setting the specific time based on a time interval between occurrences of a zero crossing time is caused.

When it is determined that the acceleration Ai is equal to or higher than the first predetermined acceleration Amax, the current limit value Li is decreased by ΔL1 (> 0) in S504. This processing serves to reduce the acceleration of the brushless motor 2 ,

If it is determined that the acceleration Ai is lower than a first specific acceleration Amax, it is determined in S506 whether or not it is equal to or less than a second specific acceleration Amin. The second specific acceleration Amin is set according to too low acceleration (too high deceleration) at which deterioration of the accuracy of setting of a certain timing is caused on the basis of a time interval between occurrences of a zero-crossing timing. When it is determined that the acceleration is equal to or smaller than the second specific acceleration Amin, the current limit Li is increased by ΔL2 (> 0) in S508. This processing serves to reduce the absolute value of the acceleration of the brushless motor 2 ,

In this embodiment The present invention will provide the following processing in one Situation carried out in which the absolute value of an acceleration Ai is excessively large and hence the accuracy of specifying a specific time is significantly deteriorated. A processing for reducing the Absolute value of acceleration is executed. The accuracy of a Setting a specific time is thereby improved. This setting is especially effective in the following cases: cases in which an excitement mainly through a 120 ° excitation process accomplished is at a speed as in this embodiment of the present invention Invention easy to control.

More specifically, acceleration is determined by a load applied to the output shaft of the brushless motor 2 and the like, and varies from situation to situation. For example, when the viscosity of fuel in a fuel pump is low, the load applied to the output shaft of the brushless motor 2 is exercised, low. Therefore, there is a possibility that acceleration at the time of startup or the like becomes too high.

If A measure carried out, To prevent acceleration in a situation too becomes high, in which the viscosity of fuel is low, the following is done. In a situation where the viscosity of fuel is high, is a start time extended. Since there is a requirement that a start time should be shortened, or still other similar ones reasons There, it is difficult to prevent an acceleration gets too high.

If the current supplied to the switching elements SW1 to SW6, exceeds a current limit, PWM control instead of turning on the switching elements SW1 to SW6 for a duration of 120 °. To This time, the state of the switching elements SW1 to SW6 is between the ON state and the OFF state in a period of 120 ° from a switched on at a specific time. At this time, the time is right to which the switching elements SW1 to SW6 to the ON state be brought first, or the time to which they come to the last time it is brought out, not always with at the specific time.

If the switching elements SW1 to SW6 are turned on and off by PWM control Become a power over the diodes D1 to D6 supplied. As a result, a duration for which are the terminal voltage Vu, Vv, Vw and the reference voltage Vref do not match, newly created. Therefore, a masking duration is set additionally.

In this embodiment According to the present invention, the current limit can only be changed when acceleration is positive and too high.

• (7) Ein Stromgrenzwert Li wird gemäß einer Beschleunigung veränderlich festgelegt. Daher kann eine Erhöhung des Absolutwerts einer Beschleunigung in dem Ausmaß unterdrückt werden, dass die Genauigkeit einer Zeit, die für ein Berechnen auf der Grundlage eines Zeitintervalls zwischen Auftrittsvorgängen eines Nulldurchgangszeitpunkts erforderlich ist, nicht übermäßig verschlechtert wird. Aus diesem Grund kann ein bestimmter Zeitpunkt unter Verwendung einer gemeinsamen Information genau festgelegt werden, die eine Änderung einer Drehzahl betrifft.

According to this eighteenth embodiment of the present invention, the following advantages can be provided in addition to the advantages (1) to (4) of the fifteenth embodiment of the present invention.

• (7) A current limit Li is variably set according to an acceleration. Therefore, an increase in the absolute value of acceleration can be suppressed to the extent that the accuracy of a time required for calculation based on a time interval between occurrences of a zero-crossing timing is not excessively deteriorated. For this reason, a specific time can be accurately determined by using a common information that has one Change of a speed concerns.

Nineteenth embodiment

The induced voltage of the brushless motor 2 will be in connection with a rotation of the brushless motor 2 generated. Because of this, when the brushless motor 2 is started in a stop state, a switching operation based on the induced voltage is not performed. To cope with it, the measure is taken in 64A is shown, normally performed when the brushless motor 2 is started. The angle (electrical angle) of the rotor is determined by passing a current from a certain phase to another specific phase, that is, positioning processing is carried out. In the example that is in 55 is shown, the switching element SW1 of the high-side branch and the switching element SW6 of the low-side branch are brought to the ON state, thereby passing a current from the U phase to the W phase to the angle of the rotor of the brushless motor 2 set. If the angle of the runner is determined by this processing, a long time may be required for the runner's angle to settle down. Because of this, there is a possibility that a starting time of the brushless motor 2 is extended.

In the nineteenth embodiment of the present invention, a processing executed in 64B is shown. That is, a switching element of one phase of the branches of the high side of the brushless motor 2 and switching elements of two phases of the low side branches are brought to the ON state to conduct a current of one phase to two phases (one-phase / two-phase excitation). The angle of the rotor of the brushless motor 2 is thereby set to a predetermined angle. 64B is an example of cases in which the following processing is performed. The switching element SW1 of the high side branch and the switching elements SW4, SW6 of the low side branches are brought to the ON state, and the current is thereby conducted from the U phase to the V phase and the W phase. Therefore, such a force is applied to the deviation of the angle of the rotor of the brushless motor 2 to decrease from a predetermined angle; Therefore, a time that it takes for the angle of the runner to settle at the predetermined angle can be shortened.

65A represents the manner in which the angle of the rotor is calmed by positioning processing, which in 64A is shown, and 65B FIG. 5 illustrates the manner in which the angle of the rotor is calmed by the positioning processing incorporated in FIG 64B is shown. In these figures, Iu, Iv, Iw indicate phase strings, and Im and Ip indicate a motor phase current and a power source current. According to the positioning processing in this embodiment of the present invention as shown in FIG 65B is shown, the time it takes for the angle of the rotor to settle down is short, and a fluctuation of the currents having a correlation with a fluctuation of the angle of the rotor is terminated early.

Setting a predetermined angle may be performed as shown in FIG 66 representing the relationship between the predetermined angle and the start time. In this example, an angle of 0 ° is defined as an angle at which it is assumed that the rotation angle is calmed when the initial state of a shift operation continues when the brushless motor 2 is started. This figure represents a start time (duration) with respect to a case in which a current corresponding to the brushless motor 2 is supplied, is not limited in the positioning processing and in a case where the current is limited.

When the predetermined angle is set in a range from an angle delayed by 150 ° to an angle advancing by 30 °, the engine may preferably be started. This refers to the fact that there is a relationship in 67 is shown between a torque that is generated when the switching operation in connection with a run-up of the brushless motor 2 is started, and gives the predetermined angle. In this example, a torque that has a positive value refers to a torque that is the brushless motor 2 turns in the positive direction. In the area on the advancing side is a torque that is the brushless motor 2 turns backwards, generates. Therefore, the brushless motor tends 2 to be turned backwards. The output torque is maximized at an angle that is delayed by 60 °. For this reason, the following can be realized in the vicinity of an angle delayed by 60 °. As it is in 66 In particular, regardless of whether the current supplied to the brushless motor is limited in positioning processing or not, the start time may be shortened. In this embodiment of the present invention, therefore, the predetermined angle is set in the vicinity of the angle which is delayed by 60 °.

68 Provides processing for starting the brushless motor 2 in this embodiment of the present invention. This processing is triggered by turning on an ignition switch (IG-ON) and from the drive control circuit 220 executed.

When the ignition switch is turned on, positioning processing of the one-phase / two-phase excitation is performed in S510. In S512, it is checked whether or not a predetermined time Tdi has elapsed after the start of positioning processing. When the predetermined time Tdi is set to a time equal to or longer than a time in which it is assumed that the angle of the rotor calms down to the predetermined angle by one-phase / two-phase excitation, and is as short as possible. When the predetermined time Tdi has elapsed, the positioning processing in S514 is finished. That is, the ON operation for the one phase of the high side branch and the two phases of the low side branches is stopped, and these phases are changed to the OFF operation state. In S516 becomes the brushless motor 2 started. The processing of S514 is referred to as the processing of changing the state of a switching operation for the previous one-phase / two-phase energization to the initial state of a switching operation in connection with a booster-motor start-up 2 be executed. When the processing of S516 is finished, this series of processing is completed once.

• (8) Vor einem Hochlaufen des bürstenlosen Motors 2 wird ein Strom von einer Phase zu zwei anderen Phasen des bürstenlosen Motors 2 zugeführt, um den Winkel des Läufers des bürstenlosen Motors 2 auf einen vorbestimmten Winkel festzulegen. Daher kann eine Zeit, die es dauert, damit der Winkel auf dem vorbestimmten Winkel beruhigt wird, verkürzt werden, und kann daher die Startzeit des bürstenlosen Motors 2 verkürzt werden.
• (9) Der vorbestimmte Winkel wird in der Nähe eines Winkels festgelegt, der um 60° von dem beruhigten Wert des Läufers des bürstenlosen Motors 2 verzögert ist, der angenommen wird, wenn der Anfangszustand eines Schaltbetriebs in Verbindung mit dem Hochlaufen des bürstenlosen Motors 2 fortgesetzt wird. Daher kann ein großes Drehmoment in der positiven Richtung von dem ersten Schaltbetrieb erzeugt werden und kann folglich die Startzeit verkürzt werden.

According to the nineteenth embodiment of the present invention, the following advantages can be provided in addition to the advantages (1) to (4) of the fifteenth embodiment of the present invention.

• (8) Before starting up the brushless motor 2 becomes a current from one phase to two other phases of the brushless motor 2 fed to the angle of the rotor of the brushless motor 2 set to a predetermined angle. Therefore, a time it takes for the angle to be calmed at the predetermined angle can be shortened, and therefore, the start time of the brushless motor 2 be shortened.
• (9) The predetermined angle is set in the vicinity of an angle which is 60 ° from the calmed value of the rotor of the brushless motor 2 is delayed, which is assumed when the initial state of a switching operation in connection with the run-up of the brushless motor 2 will continue. Therefore, a large torque can be generated in the positive direction from the first shift operation, and thus the start time can be shortened.

Twentieth Embodiment

When the preceding positioning processing is performed by the one-phase / two-phase excitation, a torque need not be the one-phase / two-phase excitation depending on the stop position of the brushless motor 2 are generated before the positioning processing is started. More specifically, how it looks 67 it can be seen that the torque is not generated when the predetermined angle and the stop position are shifted from each other by 180 °. In this case, therefore, there is a possibility that positioning can not be properly performed by the preceding processing.

Thus, in a twentieth embodiment of the present invention, positioning processing by performing one-phase / two-phase excitation is achieved twice. 69 Provides processing for starting the brushless motor 2 in this embodiment of the present invention. This processing is triggered by turning on the ignition switch and the drive control circuit 220 executed.

This series of processing is performed as follows. In S520, preliminary positioning processing is performed by one-phase / two-phase excitation. At this time, the angle of the rotor is set at an angle different from the predetermined angle. When a predetermined time Tdi has elapsed (S522: YES), the preliminary positioning processing is ended in S524. In S526, final positioning processing by the one-phase / two-phase excitation is subsequently performed. That is, this positioning processing is performed to calm the angle of the rotor to the predetermined angle. At this time, the predetermined angle is set as in the fifteenth embodiment of the present invention. That is, it is set to the angle near an angle delayed by 60 ° from the settled value of the angle of the rotor obtained when the initial state of a switching operation in connection with the running up of the brushless motor 2 will continue. The predetermined angle is set such that the angular difference between the angle at which the runner is set by the previous temporary positioning processing and the angle at which the runner is set by the final positioning processing is greater than 0 ° and less than 180 °. It is more desirably set so that the angular difference is substantially 60 °. When the time of the final positioning processing becomes equal to the predetermined time Tdi, the final positioning processing is ended in S530. In S532 becomes the brushless motor 2 started.

According to the above processing, the following can be realized even when the difference between the angle of the rotor of the brushless motor 2 and the predetermined angle is 180 °. The brushless motor 2 can be rotated by the temporary positioning processing to control the difference from the predetermined angle to less than 180 °. For this reason, the angle of the rotor of the brushless motor 2 be set to the predetermined angle without failure by the final positioning processing. If the angle of the rotor of the brushless motor 2 before a positioning processing is such an angle that the torque is not generated by the temporary positioning processing, this angle is an angle at which the torque is generated from the final positioning processing. For this reason, the rotor can be rotated to the predetermined angle by the final positioning processing.

The following can be realized by making the difference between the angle at which the rotor is set by the temporary positioning processing and the predetermined angle near 60 °. The predetermined time Tdi in S528 may be shorter than the predetermined time Tdi in S512 in FIG 68 be made. The reason is as follows. The torque in the brushless motor 2 is generated when the final positioning processing is started is increased. As a result, a time required for the angle at which the runner is set by the temporary positioning processing to transition to the predetermined angle is shortened.

• (10) Der Winkel des Läufers des bürstenlosen Motors 2 ist durch Ausführen der Einphasen/Zweiphasenerregung zwei Mal mit der geänderten festgelegten Position an einem vorbestimmten Winkel festgelegt. Deshalb kann der Winkel des Läufers an dem vorbestimmten Winkel ohne Ausfall unberücksichtigt des Winkels des Läufers des bürstenlosen Motors 2 vor der Positionierungsverarbeitung festgelegt werden.

According to the twentieth embodiment of the present invention, the following advantage can be provided in addition to the advantages (8) and (9) of the nineteenth embodiment of the present invention.

• (10) The angle of the rotor of the brushless motor 2 is set at a predetermined angle by executing the one-phase / two-phase excitation two times with the changed set position. Therefore, the angle of the rotor at the predetermined angle without failure can be disregarded from the angle of the rotor of the brushless motor 2 before positioning processing.

Twenty-first embodiment

In the twenty-first embodiment of the present invention as defined in 70 is shown, a processing for starting a brushless motor 2 triggered by turning on the ignition switch and the Ansteuer- control circuit 220 executed. This series of processing is similar but different from that of the nineteenth embodiment of the present invention ( 68 ) in S540 and S542.

When it determines in S512 that the predetermined time Tdis has elapsed, all-phase short-circuit processing in S540 is performed. In this processing, the switching elements SW1, SW3, SW5 of the high-side branches or the switching elements SW2, SW4, SW6 of the low-side branches are all turned on. When the all-phase short-circuit processing is performed for a predetermined time Ts (S542: YES), the brushless motor becomes 2 started.

According to the all-phase short-circuit processing, a current is supplied through the brushless motor 2 by the induced voltage in connection with a rotation of the brushless motor 2 fed. This current is attenuated by the resistance value of the current passage and the like. In other words, a rotational energy is attenuated. For this reason, the brushless motor 2 be stopped quickly at the predetermined angle. The predetermined time Ts is set to a time satisfying the following conditions: the time should be equal to or longer than a time in which a vibration from the all-phase short-circuit processing is so attenuated that the angle of the rotor of the brushless motor 2 settles at the predetermined angle and the engine is substantially stopped; and the time should be as short as possible.

According to this processing, it is not necessary to determine the predetermined time Tdis in S512 according to a time it takes for the angle of the rotor of the brushless motor to be long 2 Calms at the predetermined angle. The predetermined time may be determined in accordance with a time required to cause the angle of the rotor to transition to a predetermined angle. For this reason the predetermined Tdis may be shorter than the predetermined time Tdi in S512 in FIG 68 be made.

• (11) Nach der Verarbeitung durch die Einphasen/Zweiphasenerregung werden alle der Phasen des bürstenlosen Motors 2 kurzgeschlossen. Daher kann eine Zeit, die es dauert, damit der Winkel des Läufers sich an dem vorbestimmten Winkel beruhigt, weiter verkürzt werden.

According to the twenty-first embodiment of the present invention, the following advantage can be provided in addition to the advantages (8) and (9) of the nineteenth embodiment of the present invention.

• (11) After processing by the single-phase / two-phase excitation, all of the phases of the brushless motor become 2 shorted. Therefore, a time that it takes for the angle of the rotor to settle at the predetermined angle can be further shortened.

Twenty-second embodiment

71 Provides processing for starting the brushless motor 2 according to the twenty-second embodiment of the present invention. This processing is triggered by turning on the ignition switch and the drive control circuit 220 executed. This processing is similar but different in S544 and S546 to the processing of the twentieth embodiment of the present invention ( 69 ).

This series of processing is performed as follows. All-phase short-circuit processing is executed (S544 and S546) when the following processing is performed. It is executed when the one-phase / two-phase energization is performed for the preliminary positioning processing for the predetermined time Tdis (S522: YES). Further, it is executed when the one-phase / two-phase energization is executed for the final positioning processing for the predetermined time Tdis (S528: YES). Therefore, the time required for positioning can be compared with the twentieth embodiment of the present invention ( 69 ) be shortened.

Twenty-third embodiment

The twenty-third embodiment of the present invention is an embodiment of the twentieth embodiment of the present invention ( 69 ).

If the current supplied to the switching elements SW1 to SW6 equal to or greater than the predetermined value in the 120 ° excitation control PWM control is performed to restrict it. In this embodiment According to the present invention, PWM control is also carried out when the current supplied to the switching elements SW1 to SW6 equal to or greater than is a predetermined value in the positioning processing.

72 FIG. 15 illustrates processing in a current limit control in this twenty-third embodiment of the present invention. This processing is performed by the drive control circuit 220 for example, repeatedly executed in a predetermined cycle.

In S550 it will check if the Positioning processing is in progress or not. When it is determined that the positioning processing running, it is determined in S552 whether the value of a current Ih of the switching elements SW1, SW3, SW5 of the high side branches equal to or greater than a threshold is Ith or not. This processing is used for Determining if the value of a current Ih during the positioning processing is too big or not. When it is determined that the magnitude of a current is equal to Ihs the bigger one than that Threshold current Ith, all of the switching elements SW2, SW4, SW6 of the low side branches off in S554. When in S550 or S552 a negative determination is made or if the processing S554 is finished, this series of processing is completed once.

In this twenty-third embodiment According to the present invention, the current can be switched off by switching off the Switching elements of the branches of the high side are limited. In this In the case, however, processing is performed to supply the current according to, for example to limit its whether the sum of the values of currents, that go through the switching elements of the branches of the low side, equal to or greater than one predetermined value is or not.

• (12) Wenn der Wert eines Stroms, der dem bürstenlosen Motor 2 zugeführt wird, gleich oder größer als der vorbestimmte Wert aufgrund einer Einphasen/Zweiphasenerregung einer Positionierungsverarbeitung ist, wird der Wert einer Erregung beschränkt. Daher kann verhindert werden, dass eine Energieaufnahme einer Positionierungsverarbeitung übermäßig groß wird. Weiterhin ist es möglich, zu verhindern, dass ein Drehmoment, das von einem Erregungsbetrieb zum Positionieren erzeugt wird, zu hoch wird, und eine Zeit zu verkürzen, die es dauert, damit sich der Drehwinkel beruhigt.

According to the twenty-third embodiment of the present invention, the following advantage (12) can be provided in addition to the advantages (8) and (9).

• (12) If the value of a current, the brushless motor 2 is supplied, equal to or greater than the predetermined value due to a one-phase / two-phase excitation of a positioning processing, the value of an excitation is limited. Therefore, it can be prevented that an energy consumption of a Positioning processing becomes excessively large. Furthermore, it is possible to prevent a torque generated by an energizing operation for positioning from becoming too high and shortening a time it takes for the rotation angle to settle down.

Twenty-fourth embodiment

73 represents a processing that starts a processing for starting the brushless motor 2 in the twenty-fourth embodiment of the present invention. This processing is triggered by turning on the ignition switch and the drive control circuit 220 for example, repeatedly executed in a predetermined cycle.

In this processing, if a duration for which the voltage VB of the battery 214 is equal to or higher than a certain voltage Vth for a predetermined time Tv or longer (S560: YES), the following processing is executed. Various parameters for controlling the brushless motor 2 in the drive control circuit 220 are initialized (S562) and processing, which is a run-up of the brushless motor 2 is associated with is executed (S564). That is, the processing is performed in 68 is shown.

The determined voltage Vth is set to a value obtained by adding a predetermined limit to a voltage required for the operation of the drive control circuit 220 stabilized. When the ramp-up of the brushless motor 2 is limited to a state in which the voltage of the battery 214 is equal to or higher than the predetermined voltage Vth lasts for a predetermined time Tv or longer, the following can be realized. As it is in 74 is shown, the brushless motor 2 to be started when the voltage of the battery 214 has stabilized.

The reason why S562 is initialized is as follows. Because the voltage applied to the drive control circuit 220 is applied, drops once, there is a possibility that the reliability of the parameters deteriorated in the drive control circuit 220 be used.

• (13) Der bürstenlose Motor 2 wird beschränkt, gestartet zu werden, bis die Spannung des bürstenlosen Motors 2 gleich oder höher als eine vorbestimmte bestimmte Spannung wird. Daher ist es möglich, ein Ereignis zu verhindern, dass der Betrieb der Ansteuer-Steuerschaltung 220 destabilisiert wird und der Drehzustand des bürstenlosen Motors 2 außer Kontrolle gerät und dergleichen. Aus diesem Grund kann eine Situation, in welcher die Drehung des bürstenlosen Motors 2 gestoppt wird und dann eine Startverarbeitung erneut ausgeführt wird, vermieden werden.
• (14) Der bürstenlose Motor 2 wird beschränkt, gestartet zu werden, bis eine Dauer, für welche die Spannung der Batterie 214 gleich oder höher als eine bestimmte Spannung ist, für eine vorbestimmte Zeit Tv fortbesteht. Daher kann der Motor mit der stabilisierten Spannung der Batterie 214 gestartet werden.

According to the twenty-fourth embodiment of the present invention, the following advantages can be provided in addition to the advantages (8) and (9) of the twentieth embodiment of the present invention.

• (13) The brushless motor 2 is limited to be started until the voltage of the brushless motor 2 becomes equal to or higher than a predetermined certain voltage. Therefore, it is possible to prevent an event that the operation of the drive control circuit 220 is destabilized and the rotational state of the brushless motor 2 out of control and the like. For this reason, a situation in which the rotation of the brushless motor 2 is stopped and then startup processing is executed again, be avoided.
• (14) The brushless motor 2 is limited to be started until a duration for which the voltage of the battery 214 is equal to or higher than a certain voltage Tv continues for a predetermined time. Therefore, the motor with the stabilized voltage of the battery 214 to be started.

The fifteenth to twenty-fourth embodiments of the present invention as described below.

The Embodiment regarding of the nineteenth embodiment of the present invention, which in the twenty-third embodiment of the present invention may be applied to the twenty-first to twenty-second embodiments of the present invention. The design in terms of of the nineteenth embodiment of the present invention, that in the twenty-fourth embodiment The present invention can be applied to the fifteenth to nineteenth and twenty-second to twenty-third embodiments of the present invention.

The positioning processing need not be performed by turning on the switching element of one phase of the high side branch and the switching elements of two other phases of the low side branches to conduct a current from one phase to the other two phases. For example, it may be performed by turning on the switching elements of two phases of the high side branches and the switching element of another phase of the low side branch to conduct a current from the two phases to the one phase. If the brushless motor 2 of four or more phases, for example, positioning processing may be performed as follows. switching elements two phases of the high side branches are turned on, and further switching elements of two phases of the low side branches are turned on.

The predetermined angle is not limited to that described as an example with respect to the preceding embodiments of the present invention. In this case, it is desirable to set the predetermined angle in an angular range from an angle delayed from 180 ° to an angle advancing by 30 °. To generate a positive torque by an initial switching operation to the brushless motor 2 To start, it is desirable to set the predetermined angle on the delayed side. Further, from the viewpoint of reducing a start time irrespective of whether a current is limited or not, it is desirable to set the predetermined angle before an angle delayed by 120 °.

The The procedure for setting a specific time does not have to following procedures are: a time that depends on the occurrence of a Zero crossing instant immediately before occurrence of a particular one Time is required, according to a time interval between occurring events a zero crossing time point calculated; and the calculated time, which is required is changed according to a change a speed corrected. For example, a two-dimensional Illustration showing the relationship between a speed, an acceleration and a required time defined. Farther may be a required time t from a predetermined angle during a Duration from occurrence of a zero-crossing time to one Occurrence of a specific time, a speed Ni and a Acceleration Ai by an expression of a predetermined angle = Ni × t + (1/2) × Ai × t × t become.

The Method for extracting information that is a change of a rotational speed, from a result of detecting a Zero crossing time and a specific time point The basis of this information is not limited to those who in terms of the previous embodiments or the embodiments thereof of the present invention are. For example, a value can be expressed by the following expression is achieved as the value of the meter to set a specific time when an occurrence occurs Zero crossing time point: present value of the Counter × (present Value of counter / previous Value of the counter). A value determined by the present value of the counter - K × (present value of the counter - previous Value of the meter) can be considered the value of the counter to set a particular Time to be taken. That is, any method can be used as long as the previous information is using of three or more results of zero-crossing timing detection be extracted, and a specific date will be based set this information.

The structure of the current detection device 228 is not limited to those described as an example with respect to the preceding embodiments of the present invention. For example, it may be configured such that a short-circuit resistance between each switching element SW1, SW3, SW5 and the positive potential of the battery 214 is provided and a current supplied thereto is detected on the basis of the value of a voltage drop by the short-circuiting resistor.

The brushless motor 2 does not have to be an actuator installed in a fuel pump, and may be, for example, an actuator of a fan for cooling a radiator of a vehicle-mounted internal combustion engine. Further, it may be a motor provided in a data recorder or a reproducing apparatus installed in a vehicle navigation system or the like. That is, it may be a motor provided in a data recorder or a floppy disk reproducing apparatus such as a DVD or Digital Versatile Disc, a CD-ROM or Compact Disc Read Only Memory, and a hard disk. The lathe does not have to be a motor but can be a generator.

The power supply does not need a battery 214 but may be a generator which is constructed so that it converts the rotational energy of an internal combustion engine installed in a vehicle to electrical energy.

A previously described motor driver according to the present invention has a synchronization loss monitoring circuit that monitors rotation of a rotating machine such as a brushless DC motor to detect an indication of transition to a state of synchronization loss. When the indication is detected, an energization control circuit temporarily stops driving the lathe to bring it into a coasting state, and then introduces Control to resume driving the lathe. Furthermore, the motor drive apparatus includes an inverter and a drive control circuit that controls a switching operation of the inverter based on rotation of the rotating machine.

Claims (75)

Lathing machine drive device for a lathe ( 2 ), comprising: a synchronization loss predicting means ( 3 ) which monitors a state of rotation of a lathe and thereby detects an indication of the lathe transitioning to a state of synchronization loss; and a drive control device ( 4 ), which, when the indication is detected by the synchronization loss predicting means, temporarily stops driving the lathe to bring the lathe to a freewheeling state, and thereafter performs control to resume driving of the lathe. The lathing driving apparatus according to claim 1, wherein said synchronization loss predicting means ( 7 ) detects a rotational speed of the lathe which compares detected rotational speed with a normal rotational speed of the lathe and detects the indication when a difference between the detected rotational speed and the normal rotational speed becomes equal to or greater than a predetermined value. A lathing driving apparatus according to claim 2, wherein said synchronization loss predicting means ( 7 ) detects a duration equivalent to an electrical angle of 60 ° based on a zero-crossing timing of an induced voltage of the rotating machine and compares a length of the detected duration with a duration that is in a case of the rotating machine which is a three-phase rotating machine , is equivalent to an electrical angle of 60 ° at a normal speed. A lathing driving apparatus according to any one of claims 1 to 3, wherein the synchronization loss predicting means (16) 7 ) detects the indication when a duration for which a pattern of development of the output voltage of each phase for the lathe does not coincide with a predetermined development pattern becomes equal to or greater than a predetermined value. The lathing driving apparatus according to claim 1, wherein said synchronization loss predicting means ( 7 3), an induced voltage of the rotary machine is divided into three levels, which are a high level, a low level, and a non-energization level set between the high level and the low level, upon detection of a pattern of induced voltage development of each phase. The lathing driving apparatus according to claim 1, wherein said synchronization loss predicting means ( 7 ) detects a current supplied to the rotating machine, and detects the indication when a fluctuation of the current becomes equal to or higher than a predetermined value. A lathe driving method comprising: detecting an induced voltage generated in a winding of a lathe ( 2 ) is developed; Estimating a rotor position of the lathe on the basis of the detected induced voltage to thereby obtain an energization timing for the lathe, monitoring (S4, S5) a state of rotation of the lathe and thereby detecting an indication that the lathe is transitioning to a state of synchronization loss ; and stopping (S6) temporarily stopping driving of the lathe to bring the lathe to a freewheeling state when the indication is detected, and then performing control to resume driving of the lathe. Lathes driving method according to claim 7, where monitoring (S4, S5) show the indication by either (1) comparing one detected Rotational speed of the lathe with a normal rotational speed of the lathe, (2) Compare a length a recorded duration, the equivalent to an electrical angle of 60 °, based on the Zero crossing time point of the induced voltage of the lathe, with a duration that is equivalent to an electrical angle of 60 ° at normal speed, or (3) comparing a duration for which a pattern of evolvement an output voltage of each phase for the lathe not with matches a predetermined development pattern, equal to or greater than a predetermined value is detected. A lathing driving method according to claim 6, wherein said monitoring (S4, S5) is a pattern of Detecting an induced voltage of each phase of the rotating machine, and differentiating the induced voltage into three levels that are a high level, a low level and a non-energization level set between the high level and the low level. Lathes driving method according to claim 6, where monitoring (S4, S5) detects a current supplied to the lathe and the indication detected when a fluctuation of the current is equal to or greater than becomes a predetermined value. A lathe driving device for a lathe, comprising: a forced commutation device ( 4 ) to perform a forced commutation to start the lathe; a torque limiting device ( 4 . 5 ) for limiting a current supplied to a winding of the lathe to an upper limit level set higher than a level at which a current flows when the lathe is in a normal rotation state, when the forced commutation device performs the forced commutation. A lathe drive apparatus according to claim 11, wherein the forced commutation means ( 4 ) starts the forced commutation after positioning the rotor of the lathe; and the torque limiting device ( 4 . 5 ) energizes the winding at a phase angle leading by a predetermined value from a proper phase angle relative to the rotor position. Lathes driving method for a lathe ( 2 ), comprising: detecting an induced voltage developed in a winding of the lathe; Estimating a rotor position of the lathe on the basis of the detected induced voltage to thereby obtain an energization timing for the lathe; and limiting a current supplied to the winding of the lathe to an upper limit level set higher than a level at which a current flows in a steady rotating state of the lathe when the forcible commutation is performed to start the lathe , Lathes driving method according to claim 13, which further comprises: Starting the forced commutation, after the runner has been positioned at an initial position; and excite the winding of the lathe at a phase angle which um a predetermined value to an appropriate phase angle with respect to rotor position leading, at the same time as starting. Lathes driving method according to claim 14, wherein the predetermined value is about 30 ° of an electrical angle the lathe is. A lathe drive apparatus comprising: an energy converter circuit ( 12 ) including switching elements (SW1 to SW6) and rectifying elements (D1 to D6) connected in parallel with the respective switching elements to enable connection of a power supply ( 214 ) to a lathe ( 2 ) to control; and a drive control circuit ( 220 ) for detecting a time point at which an electrical angle of the lathe becomes equal to a predetermined electrical angle, based on an induced voltage of the lathe, and thereby a specific time point providing a basis for operation of the switching elements, wherein the drive control circuit ( 220 ): means (S10 to S24) for setting a permitted period for setting a permitted duration for which detection of the predetermined electrical angle is permitted on the basis of a detected value of a terminal voltage of the lathe; and determining means (S58, S78) for determining that a rotation state of the lathe is abnormal when the number of times that the predetermined electrical angle continuously occurs either before or after the released period becomes equal to or greater than a threshold value. Lathes drive device according to claim 16, where the time that the predetermined electrical angle occurs, a zero crossing time is when the induced voltage the lathe is equal to a reference voltage; and the Zero crossing time based on a comparison of a detected Value of the terminal voltage of the lathe with the reference voltage is detected. A lath drive device according to claim 17, wherein said drive control circuit ( 220 ) further includes means for determining whether or not the zero-crossing time has occurred before the enabled period; and the determined time is determined by assuming that a time of commencement of the enabled duration is the zero-crossing time when it is determined that the zero-crossing time has occurred before the enabled duration. A lath drive device according to claim 17, wherein said drive control circuit ( 220 ) further comprising means for determining whether the zero-crossing time has occurred before the enabled period and means for determining a most-advanced time that may be applied as the zero-crossing time before the enabled period; and the determined time is determined by assuming that the zero-crossing time is between the most advanced time to the time of commencement of the enabled duration, when it is determined that the zero-crossing time has occurred before the enabled duration. A lath drive device according to any one of claims 17 to 19, wherein said drive control circuit ( 220 ) includes means for determining whether or not the zero-crossing time occurs after the enabled period; and the determined time is determined by assuming that the time of one end of the enabled duration is the zero-crossing time when it is determined that the zero-crossing time occurs after the enabled duration. A lath drive device according to claim 20, wherein said drive control circuit ( 220 ) further includes comparator means for outputting a binary signal indicative of a result of a comparison between the terminal voltage or the reference voltage; and the determining means checks on the basis of the value of the binary signal at the time of commencement of the enabled period whether the zero-crossing time has occurred before the enabled period or not and on the basis of the presence or absence of a change of the logical value of the enabled duration whether the zero crossing time occurs after the enabled period or not. Lathes drive device according to one of claims 16-19, the facility for setting a shared Duration means for determining a time which is different from the predetermined electrical Angle to the shared duration is required, based on a rotational speed of the lathe includes that of a detected Value of a time interval between adjacent times when the predetermined electrical angle occurs, and a change the rotational speed of the lathe is determined. Lathes drive device according to claim 22, the change the speed based on a change of a detected value the voltage of the power supply is determined. A lath drive device according to claim 22, wherein said drive control circuit ( 220 ) further comprising extracting means for extracting information concerning a change in rotational speed based on the result of detecting a time at which the predetermined electrical angle occurs; and the change of the rotational speed is determined on the basis of the extracted information. A lathe drive apparatus comprising: an energy converter circuit ( 12 ) including switching elements (SW1 to SW6) and rectifying elements (D1 to D6) connected in parallel with the switching elements to connect a power supply ( 214 ) to a lathe ( 2 ) to control; and a drive control circuit ( 220 ) for detecting a time at which an electrical angle of the lathe becomes equal to a predetermined electrical angle based on an induced voltage of the lathe to thereby set a certain timing providing a basis for operation of the switching elements, wherein the driving Control circuit ( 220 ): means (S10 to S24) for setting a permitted period for setting a permitted duration for which detection of the predetermined electrical angle is enabled on the basis of a detected value of a terminal voltage of the lathe; and the means (S10 to S24) for setting a permitted duration includes means for setting a time required from the predetermined electric to the enabled duration based on a rotational speed of the lathe consisting of a detected value of a time interval between adjacent ones Time points of occurrence of the predetermined electrical angle and a change in the rotational speed of the lathe is determined. Lathes drive device according to claim 25, the change the speed by a change a detected value of the voltage of the power supply determined becomes. A lath drive device according to claim 25, wherein said drive control circuit ( 220 ) further comprising extracting means for extracting information concerning the change of the rotational speed on the basis of a result of detecting a time at which the predetermined electrical angle occurs, and determining the change of the rotational speed by the extracted information. Lathes drive device according to one of claims 25 to 27, wherein the time at which the predetermined electrical angle occurs, the zero crossing time to which the induced voltage the lathe becomes equal to a reference voltage, and the zero crossing time based on a comparison of a detected value of the terminal voltage the lathe and the reference voltage is detected. Lathes drive device according to one of claims 25 to 27, wherein the lathe is a brushless motor for a fuel pump for feeding of a fuel to an internal combustion engine that is in a motorcycle is installed. Lathes drive device according to claim 27, wherein the extracting means an acceleration of a Lathe as the information extracted, the change the speed is concerned. A lathe drive apparatus comprising: an energy converter circuit ( 12 ) including a switching element (SW1 to SW6); and a comparator ( 224 to 228 ) for comparing a terminal voltage of a lathe ( 2 ) having a reference voltage with respect to an amplitude for detecting a zero-crossing timing to which the reference voltage which is either a neutral voltage of the lathe or an equivalent thereof and an induced voltage of the lathe coincide with each other, and for operating the switching element based on the zero-crossing time ; and a correction device ( 30 ) for correcting an offset of at least one of a value of the terminal voltage to be compared by the comparing means when a rotational speed of the lathe is substantially zero and a value of the reference voltage to discriminate the values of the terminal voltage and the reference voltage. Lathing machine drive device, wherein the lathe is a multi-phase lathe, and the correction device ( 30 ) performs an offset correction such that a result of comparison by the comparison means when the rotational speed of the multiphase rotary machine is substantially zero becomes the same for all phases. Lathing machine drive device according to claim 31, wherein the correction device ( 30 ) a resistive element ( 30 ) including a signal wire for inputting an object to be corrected into the comparing means having a predetermined potential. Lathes drive device according to claim 33, wherein the object to be corrected is the reference voltage. A lathe drive apparatus according to claim 34, wherein the lathe ( 2 ) is a multi-phase lathe; and the reference voltage is a virtual zero point voltage obtained by dividing the terminal voltage of each phase of the multiphase rotary machine by resistive elements ( 30 , RU, RV, RW) is generated. A lathing drive apparatus according to claim 31, wherein the comparison means ( 224 to 228 ) a differential amplifier circuit ( 240 ) and an output circuit ( 250 ), and the correction device ( 30 ) includes a pair of elements in the differential amplifier circuit, each corresponding to a pair of input terminals of the comparator, the pair of elements being asymmetrical with each other. A lathe drive apparatus according to claim 36, wherein the lathe ( 2 ) is a multi-phase lathe; and the differential amplifier circuit ( 240 ) for each phase and in a single structure for each phase. Lathes drive device according to one of claims 31 to 37, which further includes means for starting an operation the switching element for starting the lathe at a stop based on a detected value of the zero crossing time having. A lathe drive method comprising: to compare a terminal voltage of a lathe with a reference voltage in terms of an amplitude to detect a zero-crossing timing the reference voltage, which is either a zero point voltage the lathe or equivalent and an induced voltage of the lathe coincide with each other; Operate the switching element based on the zero-crossing timing; and Correcting an offset of at least one of a value the terminal voltage, which is to compare when a speed the lathe is substantially zero, and a value of the reference voltage, to distinguish the values of the terminal voltage and the reference voltage. Lathes driving method, wherein the Lathe is a multi-phase lathe; and correcting an offset performs an offset correction such that a result of a comparison, which is performed when the speed the multiphase lathe is essentially zero, for all phases the same will be. Lathing machine drive device for a lathe ( 10 ) comprising: an energy converter circuit ( 12 ) including switching elements (SW1 to SW6) each having a positive pole and a negative pole of a power supply ( 214 ) to the lathe on the basis of a zero-crossing timing at which an induced voltage of the lathe becomes equal to a reference voltage; and a comparator ( 220 ) for individually comparing a terminal voltage of each phase of the lathe with the reference voltage; and a detection device ( 220 ) for detecting information concerning an electrical angle of the lathe based on a result of comparison by the comparing means assumed when the zero-crossing timing occurs in a present operating state of the switching elements and an actual result of comparison with respect to each phase. A lathing drive apparatus according to claim 41, wherein the detection means ( 220 ) includes means for detecting the zero-crossing time based on a match between an assumed result of a comparison and the actual result of a comparison with respect to at least one phase. A lathing drive apparatus according to claim 42, wherein said detection means ( 220 ) includes means for detecting the zero-crossing time based on a correspondence between the presumed result of a comparison and the actual result of a comparison with respect to all of the phases. A lathe drive device according to claim 42 or 43, further comprising a locking device ( 220 ) for setting a specific time which provides a basis for changing the operating state of the switching elements based on the zero-crossing timing. Lathes drive device according to claim 44, wherein the specific time and the zero-crossing time in a one-to-one correspondence be brought to each other. The lathing drive apparatus of claim 45, further comprising: an establishment ( 220 ) for switching between an ON operation and an OFF operation of the switching element connected to one of the positive pole and the negative pole during a period of enabling an ON operation for each switching element determined by the determined timing; wherein the energy converter circuit ( 12 ) includes a rectifier means (D1 to D6) connected in parallel with each of the switching elements, and the reference voltage is set to either a zero point voltage of the lathe or an equivalent thereof. Lathing machine drive device according to one of claims 41 to 43, wherein the detection device ( 220 ) an abnormality determination means for determining the rotational state of the lathe to be abnormal based on a mismatch between the result of comparison by the comparing means assumed when the zero-crossing timing occurs in the present operating state of the switching elements and the actual result of comparison with respect to at least one phase includes. A lathing drive apparatus according to claim 47, wherein said abnormality determination means ( 220 , S366) means for determining an existence of an abnormality that the lathe rotates backward based on a correspondence between what is obtained by temporally reversing the time series pattern of the result of comparison by the comparing means, from the time series pattern of the operating state of the switching elements is assumed, and includes the actual time sequence pattern. A lathe drive apparatus according to claim 47, further comprising: a stop means (12). 220 , S368) for forcibly stopping the rotation of the lathe when the rotational state of the lathe is determined to be abnormal; and a starting device ( 220 , S372) for restarting the lathe after stopping the lathe by the stopper. A lathe drive apparatus according to claim 49, wherein the stop means ( 220 , S372) forcibly stops the rotation of the lathe by forming a conduction of all the phases of the lathe with either the positive or the negative pole. A lathing drive apparatus according to claim 46, further comprising separation detection means (14). 220 , S390) for detecting presence / absence of disconnection of a phase line of the lathe based on presence / absence of reversal of the result of comparison of the comparing means via a phase connected to a switching element of the switching elements set to an ON state is, in a duration of processing of the operating device has. A lathe drive apparatus comprising: an energy converter circuit ( 12 ) having switching elements (SW1 to SW6) each having a positive pole and a negative pole of a power supply ( 214 ) connect to each phase of a lathe and thereby control the lathe; an anomaly determination device ( 220 , S366) for determining whether an abnormality exists in a rotating state of the rotating machine based on a detected value of an induced voltage of the rotating machine; a stop device ( 220 , S368) for directing all of the phases of the lathe with either the positive pole or the negative pole when it is determined that the anomaly is present, thereby forcibly stopping the rotation of the lathe; and a starting device ( 220 , S372) for restarting the lathe after stopping processing by the stopper. A lathe drive apparatus comprising: an energy converter circuit ( 12 ) including switching elements (SW1 to SW6) and rectifying means (D1 to D6) connected in parallel with the switching elements; a comparison device ( 220 ) for comparing a terminal voltage of the lathe and any one of a zero point voltage of the lathe and an equivalent thereof with respect to each phase of the lathe; and a separation detection device ( 220 , S390) for detecting presence / absence that is, a separation on the basis of presence / absence of reversal of the result of comparison of the comparison means corresponding to at least one phase at the time of turning off the switching element making any of the one of input terminals and the lathe under the condition that one phase of phases of the rotating machine is made conductive to one of a pair of input terminals of the power converter circuit and at least one other phase other than the one phase is rendered conductive to the other of the pair of input terminals. The lathe drive apparatus according to claim 53, further comprising: a fixing device ( 220 ) for setting a specific time as a reference for switching operating states of the switching elements based on a zero-crossing timing to which an induced voltage of the rotating machine coincides with the reference voltage; and an operating facility ( 220 ) for repeating ON / OFF operations of the switching element, which is in a duration for enabling an ON operation, in the duration for enabling ON operation of each switching element defined by the designated time when a phase current of the lathe exceeds a threshold. A lathe drive apparatus according to claim 53 or 54, wherein the lathe ( 2 ) is a brushless motor for a fuel pump for supplying a fuel to an internal combustion engine installed in a motorcycle. A lathe drive apparatus comprising: an energy converter circuit ( 12 ), which includes switching elements (SW1 to SW6) operable to drive a lathe ( 2 ) to control; an extraction device ( 220 ) for extracting information corresponding to a change in rotational speed of the lathe from a result of detecting a zero-crossing timing at which an induced voltage of the lathe becomes equal to a reference voltage; and a fixing device ( 220 ) for calculating a time required from occurrence of the zero-crossing timing to occurrence of a certain timing providing a basis for operation of the switching elements based on a time interval between occurrences of the zero-crossing timing, the setting means determining the determined timing at the time Basis of the information concerning the change of the speed, variable sets. A lathing drive apparatus according to claim 56, wherein said fixing means ( 220 ) makes the given time more advanced as the speed increases more. A lathing drive apparatus according to claim 56 or 57, wherein said fixing means ( 220 ) includes: means for calculating a required time for calculating a required time from occurrence of a zero-crossing timing immediately before occurrence of the determined timing to occurrence of the predetermined timing from a time interval between occurrences of the zero-crossing times; and a correcting means for correcting the required time based on the information. A lathe drive apparatus according to claim 56 or 57, further comprising: a comparator (16). 220 ) for comparing a terminal voltage of the lathe with a reference voltage with respect to an amplitude; and a locking device ( 220 ) for inhibiting a comparison by the comparing means for a predetermined duration of occurrence of the zero-crossing timing, the power converting circuit (15) 12 ) includes rectifying means (D1 to D6) connected in series with each of the switching elements, the zero-crossing timing being detected as a timing of reversing a result of comparison by the comparing means, and the extracting means (13) 220 ) the information is detected up to when the zero-crossing timing occurs based on a time from when lockout by the lockout device is canceled. A lathe driving apparatus according to claim 56 or 57, wherein said extracting means (16) comprises 220 ) as the information, an acceleration of the rotating machine based on a plurality of values with respect to time intervals between occurrences of the zero-crossing timing calculated; and the fixing device ( 220 ) sets the specific time using the acceleration as the information. A lathe drive apparatus according to claim 56 or 57, further comprising: a comparator (16). 220 ) for comparing a terminal voltage of the lathe with a reference voltage with respect to an amplitude; and a locking device ( 220 ) for inhibiting a comparison by the comparing means for a predetermined duration of occurrence of the zero-crossing timing, the power converting circuit (15) 12 ) includes rectifying means (D1 to D6) connected in parallel to each of the switching elements, the zero-crossing timing being determined as a timing of reversing a result of comparison by the comparing means, the required time is a time immediately from the occurrence of the zero-crossing timing prior to the occurrence of the particular time, and an estimator ( 220 ) for estimating a time elapsed from the occurrence of a zero-crossing timing immediately before the present time on the basis of the induced voltage when the occurrence of the zero-crossing timing immediately before has already ended when a lock by the inhibitor is canceled. A lathing driving apparatus according to claim 56 or 57, further comprising: an acceleration calculating means (10). 220 , S500) for calculating an acceleration of the rotating machine based on a plurality of values regarding time intervals between occurrences of the zero-crossing timing; and a limiting device ( 220 , S504, S508) for limiting an amount of excitation according to the calculated acceleration. A lathing drive apparatus according to claim 56 or 57, wherein said fixing means ( 220 ) sets the specific time on the basis of the extracted information when the lathe is started. A lathing drive apparatus according to claim 56 or 57, further comprising: a positioning device ( 220 ) for directing a current from one part of phases to another part of phases of the lathe before the lathe is started, thereby setting a rotation angle of the lathe to a predetermined angle, at least one of the one part of phases and the other part of phases includes a plurality of phases. A lathe drive apparatus comprising: an energy converter circuit ( 12 ), which includes switching elements (SW1 to SW6) operable to drive a lathe ( 2 ), wherein the switching elements are operated by determining a time required from an occurrence of a zero-crossing timing at which an induced voltage of the rotating machine becomes equal to a reference voltage to an occurrence of a certain timing, which is a basis for an operation of Provides switching elements based on a time interval between occurrences of the zero crossing time point; and an acceleration calculating device ( 200 , S500) for calculating an acceleration of the rotating machine based on a plurality of values regarding time intervals between occurrences of the zero-crossing timing; and a limiting device ( 200 , S504, S508) for limiting an amount of excitation according to the calculated acceleration. A lathe drive apparatus comprising: an energy converter circuit ( 12 ), which includes switching elements (SW1 to SW6), which is an output signal of a lathe ( 2 ) by determining a specific time for operating the lathe based on an induced voltage of the lathe; and a positioning device ( 220 ) for directing a current from one part of phases to another part of phases of the lathe before the lathe is started, thereby setting a rotation angle of the lathe to a predetermined angle, at least one of the one part of phases and the other part of phases includes a plurality of phases. Lathes drive device according to claim 66, wherein the lathe is a three-phase lathe. A lathing drive apparatus according to claim 66 or 57, wherein the positioning means ( 220 ) short-circuit means for forming conduction between one of the positive pole and the negative pole of a power supply of the lathe and all of the phases of the lathe, thereby short-circuiting all the phases of the lathe after processing conduction of the one part of phases to the lathe other part of phases of the lathe includes. A lathing drive apparatus according to claim 66 or 67, wherein the positioning means ( 220 ) performs processing of passing a current from one part of phases to another part of phases more than once, wherein one part of phases and the other part of phases has been changed at each time, and thereby the rotation angle of the lathe sets the predetermined angle. A lathe drive apparatus according to claim 69, wherein the positioning means ( 220 ) further comprises means for forming a conduction between either the positive pole or the negative pole of the power supply of the lathe and all of the phases of the lathe thereby short-circuiting all of the phases of the lathe before changing one part of phases and the other part of phases after the current has been supplied from one part of phases to the other part of phases. A lathing drive apparatus according to claim 66 or 67, further comprising limiting means (16). 220 ) for limiting the amount of excitation when a level of a current supplied to the rotary machine becomes equal to or higher than a predetermined value due to an energizing operation. A lathing drive apparatus according to claim 66 or 67, wherein the positioning means ( 220 ) sets the predetermined angle within an angular range from a first angle to a second angle, wherein the first angle is before an angle that is delayed by 180 ° from a calmed value of the rotational angle of the lathe obtained when a first shift operation is maintained in conjunction with the start of the lathe and the second angle is after an angle leading 30 ° from the settled value. The lathe drive apparatus according to claim 56, 65 or 66, further comprising: restricting means (16). 220 ) for restricting that the lathe is started until a voltage of a power supply of the lathe becomes equal to or higher than a predetermined voltage, the lathe being installed in an in-vehicle engine system. A lathing drive apparatus according to claim 73, wherein said restriction means ( 220 ), that the lathe is started until a duration for which the voltage of the power supply is equal to or higher than the predetermined voltage persists for a predetermined time. Lathes drive device according to claim 56, 65 or 66, wherein the lathe is an actuator of a fuel pump is.