JP2021093774A - Vacuum pump - Google Patents

Abstract

To provide a vacuum pump capable of preventing generation of a starting failure.SOLUTION: A vacuum pump comprises a pump rotor, a motor M driving and rotating the pump rotor, and a motor control part 400 performing sensorless driving of the motor M. The motor M is a synchronous motor where a permanent magnet is provided on a motor rotor. The motor control part 400 performs first energization control which feeds a q-axis current in a first stationary magnetic pole phase, then, performs a motor starting operation which performs second energization control which feeds a q-axis current in a second stationary magnetic pole phase different from the first stationary magnetic pole phase.SELECTED DRAWING: Figure 3

Description

The present invention relates to a vacuum pump.

Axial-flow vacuum pumps, such as turbomolecular pumps, rotate the pump rotor at high speeds for vacuum exhaust. At this time, since the lean gas is exhausted while performing compression work, the pump rotor rotates in only one direction. Therefore, the pump rotor is usually accelerated / decelerated or constantly rotated between the stationary state and the normal rotation region.

As the motor for rotationally driving the pump rotor, for example, a permanent magnet synchronous motor having a permanent magnet in the rotor is used. In a permanent magnet synchronous motor, a rotating force is obtained by using magnet torque, so it is necessary to pass a current through the stator coil at an appropriate timing to generate a rotating magnetic field. Conventionally, in order to obtain an appropriate timing, a rotation position sensor such as a Hall element is used to generate the timing. However, recently, in consideration of cost, a sensorless three-phase brushless motor that omits the rotation position sensor as described in Patent Document 1 has been proposed.

Japanese Unexamined Patent Publication No. 2002-176793

In the sensorless brushless motor described in Patent Document 1, the rotation position is acquired by using the induced voltage generated in the stator coil, and when the rotor is stopped, the stator coil rotates in a desired direction. The rotating magnetic field is generated asynchronously, and the rotation speed of the rotor is increased to the number of rotations at which the induced voltage can be obtained. However, for example, when the rotor weight is large, there is a possibility that the engine cannot be started.

The vacuum pump according to the first aspect of the present invention is a vacuum pump including a pump rotor, a motor that rotationally drives the pump rotor, and a motor control unit that drives the motor sensorlessly. The motor is permanently attached to the motor rotor. A synchronous motor provided with a magnet, the motor control unit executes a first energization control in which a q-axis current is passed in a first fixed magnetic pole phase, and then is different from the first fixed magnetic pole phase. A motor starting operation is performed to execute a second energization control in which a q-axis current is passed in the second fixed magnetic pole phase.

According to the present invention, it is possible to prevent the occurrence of a start failure in the vacuum pump.

FIG. 1 is a diagram for explaining a schematic configuration of a vacuum pump, and is a diagram showing a cross section of a pump body of a turbo molecular pump. FIG. 2 is a block diagram showing a schematic configuration of the controller. FIG. 3 is a block diagram showing a motor drive control system. FIG. 4 is a diagram showing an example of a motor rotor and a magnetic flux φ1 at the time of starting. FIG. 5 is a flowchart illustrating the starting operation process. FIG. 6 is a diagram showing generation patterns of energized states A and B. FIG. 7 is a diagram for qualitatively explaining the starting operation.

Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings. FIG. 1 is a diagram for explaining a schematic configuration of a vacuum pump, and is a diagram showing a cross section of a pump body 1 of a turbo molecular pump. The turbo molecular pump includes a pump body 1 shown in FIG. 1 and a controller (not shown) for driving the pump body 1. The turbo molecular pump shown in FIG. 1 is a magnetically levitated turbo molecular pump, and the pump body 1 includes a pump rotor 4 fastened to a shaft 5, magnetic bearings 67, 68, 69 that magnetically levitate and support the shaft 5. A motor M for rotationally driving the shaft 5 is provided. When the magnetic bearings 67, 68, 69 are not operating, the shaft 5 is supported by the mechanical bearings 66a, 66b.

The motor M is a sensorless permanent magnet synchronous motor, and the motor rotor 11 is provided with a permanent magnet. The motor stator 10 and the magnetic bearings 67 to 69 of the motor M are provided on the base 60. The pump rotor 4 is formed with a plurality of stages of rotary blades 4a and one cylindrical portion 4b. The multi-stage rotary blade 4a constitutes a turbo pump stage together with the multi-stage fixed blade 62 provided in the pump casing 61. The plurality of stages of fixed blades 62 provided alternately in the axial direction with respect to the plurality of stages of rotary blades 4a are positioned by a plurality of spacer rings 63. The cylindrical portion 4b constitutes a drag pump stage (screw groove pump) together with a pump stator 64 fixed to the base 60. Here, the screw groove is formed on the pump stator 64 side, but the screw groove may be formed on the cylindrical portion 4b side. An exhaust port 65 is provided in the exhaust port 60a formed in the base 60, and an auxiliary pump is connected to the exhaust port 65.

FIG. 2 is a block diagram showing a schematic configuration of a controller of a turbo molecular pump. The AC input from the outside is converted into a DC output (DC voltage) by the AC / DC converter 40 provided in the controller. The DC voltage output from the AC / DC converter 40 is input to the DC / DC converter 41, and the DC / DC converter 41 generates a DC voltage for the motor M and a DC voltage for the magnetic bearing.

The DC voltage for the motor M is input to the inverter 43. The DC voltage for the magnetic bearing is input to the DC power supply 42 for the magnetic bearing. As shown in FIG. 1, the magnetic bearings 67 to 69 constitute a 5-axis magnetic bearing, the radial side magnetic bearings 67 and 68 each have two pairs of electromagnets 46, and the axial side magnetic bearing 69 has one pair. It has an electromagnet 46 of. Currents are individually supplied to the five pairs of electromagnets 46, that is, the ten electromagnets 46, from the ten excitation amplifiers 45 provided for each.

The control unit 44 is a digital computer that controls a motor and a magnetic bearing. For example, a central computer (CPU), a read-only memory (ROM), a random access memory (RAM), and an input / output interface (I / O interface). ), And FPGA (Field Programmable Gate Array). The control unit 44 outputs a PWM control signal 441 for on / off control of a plurality of switching elements included in the inverter 43 to the inverter 43, and includes each excitation amplifier 45 in each excitation amplifier 45. The PWM control signal 442 for on / off control of the switching element is output. Further, as will be described later, a signal related to the motor M (signal related to the phase voltage and the phase current) 443 is input to the control unit 44, and a signal related to the magnetic bearing (excitation current signal and displacement signal) 444 are input.

FIG. 3 is a diagram showing a motor drive control system related to the motor M. The motor drive control system includes a motor control unit 400, a current detector 50, a voltage detector 51, and the like provided in the control unit 44. Although not shown, the inverter 43 includes a plurality of switching elements and a gate drive circuit for driving them on and off. The currents flowing through the U, V, and W phase coils of the motor stator 10 are detected by the current detector 50, and the current detection signal as the detection result is the rotational speed ω and magnetic pole position θ of the motor control unit 400 via the low-pass filter 409. It is input to the estimation unit (hereinafter, referred to as ω, θ estimation unit) 407. Further, the voltage of each terminal of the U, V, W phase coil and the neutral point is detected by the voltage detector 51, and the voltage detection signal as the detection result is estimated by ω, θ of the motor control unit 400 via the low-pass filter 410. It is input to the unit 407.

The ω, θ estimation unit 407 estimates the rotation speed ω and the magnetic pole position (electrical angle θ) of the motor rotor 11 based on the current detection signal and the voltage detection signal. Since the magnetic pole position of the motor rotor 11 is represented by the electric pole electric angle θ, the magnetic pole position will be referred to as the magnetic pole electric angle θ below. The calculated rotation speed ω is input to the speed control unit 401, the Id / Iq setting unit 402, and the equivalent circuit voltage conversion unit 403. Further, the calculated magnetic pole electric angle θ is input to the dq-2 phase voltage conversion unit 404.

The speed control unit 401 performs PI control (proportional control and integral control) or P control (proportional control) based on the difference between the input target rotation speed ωi and the estimated current rotation speed ω, and performs a current command. Output I. The Id / Iq setting unit 402 sets the exciting current vector component (hereinafter referred to as q-axis current) Iq and the torque current vector component (hereinafter referred to as d-axis current) Id in the rotating coordinate dq system based on the current command I. To do. As shown in FIG. 4, the d-axis of the rotating coordinate dq system is a coordinate axis whose positive direction is the north pole of the rotating motor rotor 11. The q-axis is the coordinate axis in the direction perpendicular to the d-axis with a π / 2 [rad] advance, and the direction is the counter electromotive voltage direction during normal rotation.

Generally, in motor control in a turbo molecular pump, the d-axis current Id is controlled as Id = 0. The equivalent circuit voltage conversion unit 403 is based on the rotation sensorless control parameters Ke [V0-pk / (rad / t)], r [Ω], and L [mH] stored in advance, and the following equation (1), The voltage commands Vd and Vq in the rotating coordinate dq system represented by (2) are obtained and input to the dq-2 phase voltage conversion unit 404. Here, r is the winding resistance of the entire motor wire, Ke is the counter electromotive force constant, and the angular frequency of the ω motor.
Vd = −ω × L × Iq… (1)
Vq = r × Iq + Ke × ω… (2)

The dq-2 phase voltage conversion unit 404 sets the voltage commands Vd and Vq in the rotating coordinate dq system to the fixed coordinate αβ system based on the voltage commands Vd and Vq and the magnetic pole electric angle θ input from the ω and θ estimation unit 407. It is converted into the voltage commands Vα and Vβ of. The two-phase to three-phase voltage conversion unit 405 converts the two-phase voltage commands Vα and Vβ into the three-phase voltage commands Vu, Vv and Vw. The PWM signal generation unit 406 generates a PWM control signal for turning on / off (conducting or shutting off) the switching element provided in the inverter 43 based on the three-phase voltage commands Vu, Vv, and Vw. The inverter 43 turns on and off the switching element based on the PWM control signal input from the PWM signal generation unit 406, and applies a drive voltage to the motor M.

(Conventional starting operation)
As a conventional starting operation, for example, a method is known in which a q-axis current Iq is passed in a fixed magnetic pole phase to give a trigger for the motor rotor 11 to start rotating. Generally, in a vacuum pump, during normal rotor rotation, it advances by π / 2 [rad] in the rotation direction with respect to the magnetic pole electric angle θ (magnetic pole position) and synchronizes with the phase current vector, that is, rotor rotation (rotation speed ω). A current vector (Id = 0, Iq> 0) in the q-axis direction of the dq-axis that rotates is given. However, in the case of a sensorless synchronous motor, since the counter electromotive voltage is not generated when the motor rotor 11 is stopped, the magnetic pole electric angle θ (magnetic pole position) of the motor rotor 11 cannot be detected. Therefore, at the time of starting when the magnetic pole electric angle θ (magnetic pole position) cannot be detected, a current vector oriented in a certain direction is given in the fixed coordinate αβ system to start the operation.

For example, in FIG. 4, a current vector in the α-axis positive direction (U-phase direction) is given to form a magnetic flux φ1 in the α-axis positive direction (U-phase direction) instead of a rotating magnetic flux. In FIG. 4, since the magnetic pole electric angle θ is set with reference to the U-phase direction (α-axis positive direction) (θ = 0), the q-axis current Iq at the timing when the N pole of the motor rotor 11 faces the β-axis negative direction. Will be shed. That is, assuming that the magnetic pole position of the motor rotor 11, that is, the magnetic pole electric angle θ is θ = −π / 2 [rad], the q-axis current Iq is passed at the fixed magnetic pole phase θ = −π / 2 [rad]. Become.

When such a q-axis current Iq is given, when the magnetic pole electric angle θ of the motor rotor 11 is 0 to + π [rad], a clockwise (reverse direction) torque acts on the motor rotor 11, and the magnetic pole electric angle θ becomes. In the case of 0 to −π [rad], a counterclockwise (forward rotation direction) torque acts on the motor rotor 11. Therefore, when the motor rotor is stopped as shown in FIG. 4, a counterclockwise torque acts on the motor rotor 11 to give an opportunity to start rotating in the reverse direction. However, when the motor rotor 11 is stopped at θ = 0, the north pole of the motor rotor 11 is attracted in the U-phase direction and the stopped state is maintained, so that a trigger for starting rotation cannot be given and the start fails. Will end up.

(Starting operation of this embodiment)
FIG. 5 is a flowchart illustrating a start operation process executed by the control unit 44. When the motor control is started by the motor start command, the q-axis current Iq = 10 [A] is passed at the fixed magnetic pole phase set in advance in step S10, and the timing is started. This is an operation for giving a trigger for rotation as in the conventional case. The current value of the q-axis current Iq at the time of starting is preferably the maximum current value that can be supplied to the motor M. In step S20, it is determined whether or not the rotation speed ω of the motor rotor 11 is ω = 0, that is, whether or not it is stopped, and if it is determined that ω = 0, the process proceeds to step S30. On the other hand, when it is determined that ω ≠ 0, the rotation speed ω can be estimated from the motor back electromotive voltage generated at that time, so the process proceeds to step S70 to start vector control.

When it is determined in step S20 that ω = 0 and the process proceeds to step S30, in step S30, it is determined whether or not the elapsed time from the start of timing in step S10 has passed the preset time Tset1. If it is determined in step S30 that (elapsed time) <Tset1, the process returns to step S20, and if it is determined that (elapsed time) ≥ Tset1, the process proceeds to step S40 and the q-axis current Iq is set to step S10. Switch to Iq = -10 [A] with the opposite sign.

In step S50, it is determined whether or not the elapsed time from the start of timekeeping in step S10 has passed the time Tset2. If it is determined in step S50 that (elapsed time) <Tset2, the process returns to step S20, and if it is determined that (elapsed time) ≥ Tset2, the process proceeds to step S60. In step S60, the time counting time is cleared and the process returns to step S10. That is, the q-axis current Iq is switched to Iq = 10 [A] again, and the timing is started.

In the starting process shown in FIG. 5, a pattern in which the energized state A of Iq = 10 [A] is continued for the time Tset1 and Iq until ω ≠ 0 and a trigger for rotation is given, as shown in FIG. = The pattern in which the energized state B of -10 [A] is continued for a period of time (Tset2-Tset1) is repeatedly executed. Of course, it is not necessary to repeat the process, but by repeatedly executing the energized states A and B, the start can be reliably performed. The duration of the energized state A is Tset1, and the total time of the duration of the energized state A and the duration of the energized state B is arbitrary as the set value of Tset2. For example, Tset2 = 10 seconds and Tset1 = 8 seconds. Is set as. Of course, the values of Tset1 and Tset2 are not limited to these, and for example, the length of the total time Tset2 can be set according to the size of the rotor weight, or the ratio of the durations of the energized states A and B can be changed. You may.

Next, the starting operation shown in FIG. 5 will be qualitatively described with reference to FIGS. 4, 7, and the like. The preset fixed magnetic pole phase in step S10 is arbitrary, and here, the q-axis current Iq = 10 [A] is set at the fixed magnetic pole phase such that the magnetic flux φ1 in the U-phase direction (α-axis positive direction) is formed. It shall be shed. When the motor rotor 11 is stopped in the state as shown in FIG. 4, for example, the magnetic flux φ1 causes the motor rotor 11 to rotate clockwise so that the north pole approaches the U phase. As a result, it is determined in step S20 that ω ≠ 0, and vector control is started (step S70).

On the other hand, as shown in FIG. 7, when the motor rotor 11 is stopped at the magnetic pole electric angle θ = 0, the magnetic flux formed when the q-axis current Iq is passed at the fixed magnetic pole phase θ1 = −π / 2. The direction of φ1 and the direction of the magnetic flux of the motor rotor 11 coincide with each other. Therefore, even if the q-axis current Iq = 10 [A] is passed, the torque that rotates the motor rotor 11 is not generated. Therefore, it is determined in step S20 that ω = 0, and the process proceeds to step S30. If the elapsed time is less than Tset1, the process returns from step S30 to step S20, so that the process of step S20 → step S30 is repeated, and the energized state A is maintained as in 0 seconds to Tset1 in FIG.

When the elapsed time from when Iq = 10 [A] is set in step S10 reaches Tset1, the process proceeds from step S30 to step S40, and the q-axis current Iq switches from 10 [A] to -10 [A]. Then, it changes to the energized state B as shown in FIG. In the energized state B, the q-axis current Iq having the opposite sign to that in the energized state A flows in the same fixed magnetic pole phase as in the energized state A, and as shown in FIG. 7, a magnetic flux φ2 opposite to the magnetic flux φ1 is formed. Will be done. In other words, the energized state B is a state in which the q-axis current Iq = 10 [A] flows in the fixed magnetic pole phase θ2 (θ2 = + π / 2 [rad]).

In the energized state B, the direction of the magnetic flux formed by the permanent magnet of the motor rotor 11 and the direction of the magnetic flux φ2 are opposite to each other, so that the north pole of the motor rotor 11 and the north pole on the stator side face each other and repel each other. It becomes an unstable state. As a result, the motor rotor 11 starts to move in the forward rotation direction or the reverse rotation direction, and in step S20, it is determined that ω ≠ 0 (NO), and vector control is started.

If ω ≠ 0 is not determined and the engine does not start even if Iq = -10 [A] is set in step S40, the energized state A of Iq = -10 [A] is continued (Tset2-Tset1). At a time point (when the duration from 0 seconds becomes Tset2), the process proceeds in the order of step S50 → step S60 → step S10, and the energized state A (Iq = 10 [A]) and timing are restarted again. Then, the energized state A and the energized state B are alternately and repeatedly executed until it is determined in step S20 that ω ≠ 0.

As described above, in the present embodiment, in step S10, the energized state A in which the q-axis current Iq (Iq = 10 [A]) is passed at the fixed magnetic pole phase θ1 (θ1 = −π / 2) is kept for a predetermined time (8 seconds). By maintaining), it is possible to surely obtain a state in which the magnetic pole electric angle (magnetic pole position) of the motor rotor 11 substantially matches the fixed magnetic pole phase θ1 of the q-axis current Iq. Then, by switching to the energized state B in which the q-axis current Iq (Iq = 10 [A]) flows at the fixed magnetic pole phase θ2 (θ2 = + π / 2) having a phase different from the fixed magnetic pole phase θ1, the motor rotor 11 is subjected to. Torque can be generated reliably. As a result, it is possible to prevent the failure of the rotary start in the vacuum pump.

(Modification example)
In the above-described embodiment, the sign of the current value is switched to minus from the energized state A in which the q-axis current Iq = 10 [A] flows at the fixed magnetic pole phase θ1, and the fixed magnetic pole phase θ2 is θ2 = + π / 2 and the q-axis. The energized state B in which the current Iq = 10 [A] was passed was set. However, if the motor rotor 11 that forms the energized state A and is stopped at the magnetic pole electric angle θ = 0 as shown in FIG. 7 is in the energized state B that can generate the torque that triggers the rotation, the fixed magnetic pole phase θ2 is not limited to θ2 = + π / 2. In principle, torque is generated if θ2 ≠ −π / 2, but in reality it is better to set it to about 0 ≦ θ2 ≦ π in consideration of the magnitude of the generated torque with respect to the rotor inertia. Further, the magnitude of the current value in the energized states A and B does not have to be the maximum current value that can be supplied as described above, and the current value may be different between the energized state A and the energized state B.

It will be understood by those skilled in the art that the above-described exemplary embodiments and modifications are specific examples of the following embodiments.

[1] The vacuum pump according to one aspect is a vacuum pump including a pump rotor, a motor for rotationally driving the pump rotor, and a motor control unit for driving the motor sensorlessly. The motor is a permanent magnet on the motor rotor. Is a synchronous motor provided with the above, the motor control unit executes a first energization control in which a q-axis current is passed in the first fixed magnetic pole phase, and thereafter, a second different from the first fixed magnetic pole phase. A motor starting operation is performed to execute a second energization control in which a q-axis current is passed in the fixed magnetic pole phase of 2.

For example, the energized states A and B in FIG. 7 correspond to the first energized control and the second energized control, respectively. In the energized state A, the q-axis current Iq (=) at the first fixed magnetic pole phase θ1 = −π / 2. 10 [A]) is passed, and in the energized state B, the q-axis current Iq (= 10 [A]) is passed at the second fixed magnetic pole phase θ2 = + π / 2. As a result, the magnetic pole electric angle (magnetic flux position) of the motor rotor 11 is aligned with the first fixed magnetic pole phase θ1 = −π / 2 by performing the first energization control, and the first fixed magnetic pole in the second energization control. Since the q-axis current Iq flows at the second fixed magnetic pole phase θ2 = + π / 2 having a phase different from the phase θ1 = −π / 2, the magnetic flux φ2 formed by the direction of the magnetic flux φ1 of the motor rotor 11 and the q-axis current Iq. The direction can be changed, and the occurrence of start failure can be prevented.

[2] In the vacuum pump according to the above [1], the second fixed magnetic pole phase differs from the first fixed magnetic pole phase by π [rad], and the first and second energization controls are performed. The magnitudes of the q-axis currents in are set to equal values. By setting in this way, the direction of the magnetic flux φ2 formed by the q-axis current Iq in the second energization control as shown in FIG. 7 can be reversed with respect to the magnetic flux φ1 of the motor rotor 11. As a result, the generated torque can be increased as compared with the case where the direction of the magnetic flux φ2 is other than the opposite direction.

[3] In the vacuum pump according to the above [1], the phase advance Δθ of the second fixed magnetic pole phase in the motor rotation direction is 0 [rad] <Δθ <+ π with respect to the first fixed magnetic pole phase. Set to [rad]. By setting the second fixed magnetic pole phase in this way, it is possible to generate a torque for starting the motor rotor 11 in the forward rotation direction, and it is possible to prevent the occurrence of reverse rotation.

[4] In the vacuum pump according to any one of the above [1] to [3], the current value of the q-axis current in the first and second energization control is the maximum current value that can be supplied to the motor. Is set to. As a result, the torque generated at the time of starting can be made larger, and the effect of preventing starting failure can be further improved. The maximum current value that can be supplied to the motor is both the maximum current value that can be passed through the motor M and the maximum current value that can be supplied to the motor M by the power supply (DC / DC converter 41). Means.

[5] In the vacuum pump according to the above [1], the motor control unit repeatedly executes the first energization control and the second energization control. As a result, even if the engine cannot be started by the first energization control and the second energization control, the start failure is caused by repeatedly executing the first energization control and the second energization control. It can be reliably prevented.

Although various embodiments and modifications have been described above, the present invention is not limited to these contents. Other aspects conceivable within the scope of the technical idea of the present invention are also included within the scope of the present invention.

1 ... Pump body, 4 ... Pump rotor, 10 ... Motor stator, 11 ... Motor rotor, 44 ... Control unit, 400 ... Motor control unit, A, B ... Energized state, M ... Motor

Claims (5)

With the pump rotor
A motor that rotationally drives the pump rotor and
In a vacuum pump including a motor control unit that drives the motor sensorlessly.
The motor is a synchronous motor in which a permanent magnet is provided in a motor rotor.
The motor control unit executes the first energization control in which the q-axis current flows in the first fixed magnetic pole phase, and then applies the q-axis current in the second fixed magnetic pole phase different from the first fixed magnetic pole phase. A vacuum pump that performs a motor starting operation that executes a second energization control to flow. In the vacuum pump according to claim 1,
The second fixed magnetic pole phase differs from the first fixed magnetic pole phase by π [rad], and the magnitudes of the q-axis currents in the first and second energization controls are set to equal values. ,Vacuum pump. In the vacuum pump according to claim 1,
A vacuum pump in which the phase advance Δθ of the second fixed magnetic pole phase in the motor rotation direction is set to 0 [rad] <Δθ <+ π [rad] with respect to the first fixed magnetic pole phase. In the vacuum pump according to any one of claims 1 to 3.
A vacuum pump in which the current value of the q-axis current in the first and second energization controls is set to the maximum current value that can be supplied to the motor. In the vacuum pump according to any one of claims 1 to 4.
The motor control unit is a vacuum pump that repeatedly executes a first energization control and a second energization control.

Images