JP6801481B2 - Magnetic bearing equipment and vacuum pump - Google Patents

Description

The present invention relates to a magnetic bearing device and a vacuum pump.

In a device such as a magnetic bearing type turbo molecular pump in which a rotating body (rotor) is non-contactly supported by a magnetic bearing device, the displacement between the floating position and the target position of the rotor (in order to keep the rotor floating at a predetermined target position) The magnetic attraction force (electromagnet current) of the electromagnet is feedback-controlled in real time based on the displacement). In the invention described in Patent Document 1, the displacement of the rotor is detected by using a displacement sensor arranged in the vicinity of the electromagnet, and the electromagnet current is controlled based on the detected displacement.

Japanese Unexamined Patent Publication No. 2006-308074

By the way, in the case of a radial magnetic bearing, the displacement sensor is generally arranged at a position deviated from the position of the electromagnet in the axial direction of the rotor due to the arrangement space. Therefore, if the rotor is displaced parallel to the central axis, which is the levitation target position, the displacement detected by the displacement sensor matches the displacement at the electromagnet position, but the rotor is tilted with respect to the central axis. If so, the displacement detected by the displacement sensor and the displacement at the electromagnet position do not match.

Originally, the electromagnet force applied to the rotor is preferably performed based on the displacement at the electromagnet position. However, if the displacement detection position does not match the electromagnet position as described above, the time required to return to the predetermined position will be slightly longer, or pump vibration due to noise will occur in an environment with small disturbance. There was a drawback that it became noticeable.

The magnetic bearing apparatus according to the preferred embodiment of the present invention is detected by a radial magnetic bearing electromagnet that magnetically levitates and supports the rotor shaft in the radial direction, a displacement sensor that detects the displacement of the rotor shaft in the radial direction, and the displacement sensor. A displacement conversion unit that converts the sensor position displacement into a magnet position displacement at the position of the radial magnetic bearing electromagnet, and a displacement conversion unit that controls the exciting current of the radial magnetic bearing electromagnet based on the magnet position displacement to raise the rotor shaft. Is provided with a current control unit that controls the position of the sensor, and when the value of the sensor position displacement is within a range equal to or less than a predetermined threshold value, the radial magnetic bearing electromagnet is excited based on the magnet position displacement. The current is controlled, and when the value of the sensor position displacement is not included in the range, the exciting current of the radial magnetic bearing electromagnet is controlled based on the sensor position displacement .
In a further preferred embodiment, the gain of the current controller when the value of the sensor position displacement is included in the range, than the gain of the current controller when the value of the sensor position displacement is not included in the scope Also set small.
A vacuum pump according to a preferred embodiment of the present invention includes a pump rotor, a motor that rotationally drives the pump rotor, and the magnetic bearing device that magnetically levitates and supports the rotor shaft of the pump rotor.

According to the present invention, it is possible to reduce the vibration of the magnetic bearing device.

FIG. 1 is a diagram showing a schematic configuration of a magnetic bearing type turbo molecular pump provided with a magnetic bearing device. FIG. 2 is a block diagram showing a schematic configuration of the control unit. FIG. 3 is a schematic view showing a configuration related to a magnetic bearing electromagnet for one control shaft. FIG. 4 is a diagram illustrating the characteristics of the magnetic bearing system. FIG. 5 is a block diagram showing a schematic configuration of a radial magnetic bearing control system. FIG. 6 is a diagram illustrating a conversion process related to displacement. FIG. 7 is a diagram showing a frequency distribution of current amplitude and displacement amplitude when random noise is superimposed. FIG. 8 is a block diagram showing a schematic configuration of the magnetic bearing control system according to the second embodiment. FIG. 9 is a transfer function block diagram relating to magnetic bearing control based on the current setting signal.

Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings.
− First Embodiment −
FIG. 1 is a diagram showing a schematic configuration of a magnetic bearing type turbo molecular pump provided with a magnetic bearing device. The turbo molecular pump is composed of a pump main body 1 and a control unit for driving and controlling the pump main body 1. In FIG. 1, the control unit is not shown.

The rotor shaft 5 provided in the pump rotor 3 is non-contact supported by the radial magnetic bearings 4A and 4B and the axial magnetic bearing 4C. Each of the radial magnetic bearings 4A and 4B includes four magnetic bearing electromagnets arranged in the radial direction of the rotor shaft 5. The magnetic bearing electromagnet of the axial magnetic bearing 4C is arranged so as to sandwich the thrust disk 10 fixed to the lower part of the rotor shaft 5 in the axial direction.

The displacement of the rotor shaft 5 is detected by the displacement sensor 50x1, 50y1, 50x2, 50y2 in the radial direction and the displacement sensor 51 in the axial direction. For the displacement sensor 50x1, 50y1, 50x2, 50y2, 51, an inductance type displacement sensor having a configuration in which a coil is wound around the sensor core is used.

The pump rotor 3 rotatably magnetically levitated by a magnetic bearing is rotationally driven at high speed by a motor 42. A brushless DC motor or the like is used for the motor 42. In FIG. 1, the motor 42 is schematically described, but more specifically, the portion indicated by reference numeral 42 constitutes a motor stator, and the motor rotor is provided on the rotor shaft 5 side.

A sensor target 29 is provided at the lower end of the rotor shaft 5 that is rotationally driven by the motor 42. The displacement sensor 51 in the axial direction described above is arranged at a position facing the lower surface of the sensor target 29. When the magnetic bearings are not operating, the rotor shaft 5 is supported by emergency mechanical bearings 26a, 26b.

The pump rotor 3 is formed with a plurality of stages of rotary blades 3a and a cylindrical portion 3b that form a rotating side exhaust function portion. On the other hand, on the fixed side, a fixed wing 22 and a screw stator 24, which are fixed side exhaust function portions, are provided. The multi-stage fixed blades 22 are arranged alternately with the rotary blades 3a in the axial direction. The screw stator 24 is provided on the outer peripheral side of the cylindrical portion 3b with a predetermined gap.

Each fixed wing 22 is placed on the base 20 via a spacer ring 23. When the fixing flange 21c of the pump casing 21 is fixed to the base 20 with bolts, the laminated spacer ring 23 is sandwiched between the base 20 and the pump casing 21, and the fixed wing 22 is positioned. The base 20 is provided with an exhaust port 25, and a back pump is connected to the exhaust port 25. By magnetically levitating the pump rotor 3 and driving it to rotate at high speed by the motor 42, gas molecules on the intake port 21a side are exhausted to the exhaust port 25 side.

FIG. 2 is a block diagram showing a schematic configuration of the control unit. The AC input from the outside is converted from alternating current to direct current by the DC power supply 40 provided in the control unit. The DC power supply 40 generates a power supply for the inverter 41, a power supply for the excitation amplifier 43, and a power supply for the control unit 44, respectively.

The inverter 41 that supplies electric power to the motor 42 is provided with a plurality of switching elements. The motor 42 is driven by controlling the on / off of these switching elements by the control unit 44.

As described above, the magnetic bearing that magnetically levitates and supports the rotor shaft 5 is a 5-axis control type magnetic bearing having 4 axes in the radial direction and 1 axis in the axial direction. Since a pair of magnetic bearing electromagnets is provided for each axis, 10 magnetic bearing electromagnets 45 are provided as shown in FIG. The excitation amplifier 43 that supplies electric power to the magnetic bearing electromagnet 45 is provided in each of the ten magnetic bearing electromagnets 45.

The control unit 44 that controls the drive of the motor 42 and the drive of the magnetic bearing is composed of, for example, a digital arithmetic unit such as an FPGA (Field Programmable Gate Array) and its peripheral circuits. Regarding motor control, a PWM control signal 411 for on / off control of a plurality of switching elements provided in the inverter 41 is input from the control unit 44 to the inverter 41. Further, a signal 412 relating to the phase voltage and the phase current relating to the motor 42 is input from the inverter 41 to the control unit 44.

Regarding the magnetic bearing control, a PWM gate drive signal 431 for on / off control of the switching element included in the exciting amplifier 43 is input from the control unit 44 to each exciting amplifier 43. Further, a current detection signal 432 regarding the current value of each magnetic bearing electromagnet 45 is input from each exciting amplifier 43 to the control unit 44.

A sensor circuit 33 is provided in each displacement sensor 50x1, 50y1, 50x2, 50y2, 51. A sensor carrier signal (carrier signal) 305 is input from the control unit 44 to each sensor circuit 33. A sensor signal 306 modulated by the displacement of the rotor shaft is input from each sensor circuit 33 to the control unit 44.

FIG. 3 is a schematic view showing a configuration related to the magnetic bearing electromagnet 45 for one control shaft. FIG. 3 shows the configuration of the radial magnetic bearing 4A of FIG. 1 with respect to one axis in the x direction. A pair of magnetic bearing electromagnets 45m and 45p that perform magnetic levitation support in the radial direction are arranged so as to sandwich the rotor shaft 5. J is a levitation target position when the rotor shaft 5 is magnetically levitated. Excitation amplifiers 43m and 43p are provided on the magnetic bearing electromagnets 45m and 45p. The exciting current flowing through the magnetic bearing electromagnet 45m is detected by the current sensor 46m, and the exciting current flowing through the magnetic bearing electromagnet 45p is detected by the current sensor 46p. Further, displacement sensors 50x1m and 50x1p in the radial direction are provided in association with the magnetic bearing electromagnets 45m and 45p.

As shown in FIG. 3, when the rotor shaft 5 approaches the magnetic bearing electromagnet 45p by the displacement Δdr and the gap between the magnetic bearing electromagnet 45p and the rotor shaft 5 changes, the gap change is detected by a pair of displacement sensors 50x1m and 50x1p. .. When the exciting current of the magnetic bearing electromagnet 45p is decreased and the exciting current of the magnetic bearing electromagnet 45m on the opposite side is increased, the magnet is attracted in the direction of the magnetic bearing electromagnet 45m. As a result, is controlled so that the deviation of the levitation position of the rotor shaft 5 with respect to the levitation target position J becomes small.

The exciting current flowing through the magnetic bearing electromagnets 45m and 45p includes a bias current (DC current component) for ensuring a predetermined bearing rigidity and a control current for controlling the floating position of the rotor shaft 5. .. That is, it is the control current that changes according to the floating position of the rotor shaft 5. Assuming that the levitation position of the rotor shaft 5 is the levitation target position J when the control current = 0, when the rotor shaft 5 is displaced toward the magnetic bearing electromagnet 45p as shown in FIG. 3, the magnetic bearing electromagnet The control current of 45p is + Δi, and the control current of the magnetic bearing electromagnet 45m on the opposite side is −Δi. Here, + Δi and −Δi correspond to the current fluctuations of the exciting current.

Assuming that the floating position deviation in the magnetic bearing electromagnet 45p direction at this time is the displacement Δdr, the gap between the magnetic bearing electromagnet 45p and the rotor shaft 5 is D−Δdr, and the gap between the magnetic bearing electromagnet 45m and the rotor shaft 5 is D + Δdr. Become. In this case, the force Fp in the right direction shown by the magnetic bearing electromagnet 45p and the force Fm in the left direction shown by the magnetic bearing electromagnet 45m are expressed by the following equations (1) and (2). In the equations (1) and (2), D is the gap dimension when the rotor shaft 5 is magnetically levitated to the levitation target position J, and I is the bias current (DC current component) included in the exciting current.
Fp = k ((I + Δi) / (D−Δdr)) ² … (1)
Fm = k ((I−Δi) / (D + Δdr)) ² … (2)

When the changes ΔFp and ΔFm of the forces Fp and Fm are obtained by the linear approximation of the equations (1) and (2), the equations (3) and (4) are approximately obtained. Therefore, the electromagnet force ΔF (= Fp−Fm) acting on the rotor shaft 5 is expressed by the following equation (5).
ΔFp = (2kI / D ² ) Δi + (2kI ² / D ³ ) Δdr… (3)
ΔFm = (-2kI / D ² ) Δi + (-2kI ² / D ³ ) Δdr… (4)
ΔF = ΔFp−ΔFm
= (4kI / D ² ) Δi + (4kI ² / D ³ ) Δdr… (5)

As shown in Equation (5), electromagnet force ΔF is force dependent on the control current .DELTA.i (current variation of the exciting current) in the levitation control and (4kI / D ²⁾ Δi, force dependent on displacement Δdr (4kI ² / D ³ ) It is represented by Δdr. As for the magnitude of these forces, the force caused by the current fluctuation of the exciting current is larger as shown in the equation (6).
(4kI / D ² ) | Δi |> (4kI ² / D ³ ) | Δdr |… (6)

For example, when a disturbance acts on the vacuum pump, the displacement Δdr is generated by vibrating the vacuum pump body due to the disturbance. The control unit 44 controls the rotor shaft 5 to return to the ascending target position J by passing the control currents Δi and −Δi through the magnetic bearing electromagnets 45m and 45p so as to eliminate the displacement Δdr. At this time, according to the relationship of the equation (6), the force ΔF acts on the rotor shaft 5, and the reaction acts on the pump body side. Therefore, if the rotor shaft 5 vibrates for some reason, the pump body also vibrates due to the new rotor displacement.

In the radial magnetic bearing as shown in FIG. 3, it is preferable to control the exciting currents of the magnetic bearing electromagnets 45m and 45p based on the radial displacement of the rotor shaft 5 at the electromagnet position. However, in the magnetic bearing device, it is physically difficult to arrange the displacement sensors 50x1, 50y1 and 50x2, 50y2 at the same position as the magnetic bearing electromagnet, and as shown in FIG. 3, the displacement sensor 50x1, 50y1 is placed on the magnetic bearing electromagnet 45m. , 45p is generally arranged at a position displaced in the axial direction.

Therefore, when the rotor shaft 5 is displaced parallel to the levitation target position J, the displacement at the electromagnet position and the displacement detected by the displacement sensors 50x1p and 50x1p match, but are tilted with respect to the levitation target position J. If so, the displacement at the electromagnet position and the displacement detected by the displacement sensors 50x1p and 50x1p do not match. If the displacements do not match, optimum bearing control cannot be performed, so that the time required to return the position of the rotor shaft 5 to the levitation target position J becomes longer, and the amount of work required for returning increases accordingly.

Furthermore, when the actual displacement (displacement at the electromagnet position) and the displacement detected by the displacement sensors 50x1p and 50x1p do not match, the amount of fluctuation of the exciting current during bearing control is larger than when they match. It becomes larger, and the vibration of the pump body due to the reaction of the electromagnet force ΔF becomes larger.

In the above description of the electromagnet force ΔF, the transient response when a transient disturbance from the external environment acts is described as an example, but the noise caused by the magnetic bearing system is generated when the displacement signal of the displacement sensor is sensed or excited. A similar problem occurs when the current signal is mixed during the current signal sensing required for amplifier control. Usually, in applications where weak pump vibration is a problem, for example, when used in an electron microscope, vibration generation due to the magnetic bearing system is more problematic than the influence of disturbance.

In the magnetic bearing system of a magnetic bearing turbo molecular pump, the dynamic rigidity in the low frequency band is low (high compliance), and the superimposed noise (random noise) is unique to the age difference mode in the magnetic bearing turbo molecular pump. Displacement fluctuations and current fluctuations become large in the low frequency range (several tens of Hz or less) including frequencies (see FIG. 4). Note that FIG. 4 is a diagram showing the frequency distribution of the current amplitude and the displacement amplitude when random noise is superimposed in the conventional magnetic bearing system.

Therefore, conventionally, the exciting current of the magnetic bearing electromagnet is controlled based on the displacement detected by the displacement sensor (hereinafter referred to as sensor position displacement), but in the present embodiment, the displacement at the electromagnet position (hereinafter referred to as sensor position displacement). The exciting current is controlled based on the displacement of the magnet position).

FIG. 5 is a block diagram showing a schematic configuration of a magnetic bearing control system for the radial magnetic bearings 4A and 4B shown in FIG. In FIG. 5, the magnetic bearing electromagnet 45 represents eight magnetic bearing electromagnets for four radial axes, and similarly, the displacement sensor 50, the sensor circuit 33, and the exciting amplifier 43 also correspond to the eight magnetic bearing electromagnets. Represents eight displacement sensors, sensor circuits, and exciting amplifiers, each of which is provided. The exciting current supplied from the exciting amplifier 43 to the magnetic bearing electromagnet 45 is detected by the current sensor 46.

The displacement sensor signal output from the sensor circuit 33 is input to the demodulation unit 441 of the control unit 44, and the demodulation process is performed there. From the demodulator 441, the displacement based on the displacement sensor signal, that is, the sensor position displacement Dxs1, Dys1 which is the displacement at the axially upper displacement sensor position in FIG. 1, and the sensor position displacement Dxs2 at the axially lower displacement sensor position, Dys2 is output. The displacement conversion unit 442 is based on the input sensor position displacements Dxs1, Dys1, Dxs2, Dys2, and is the displacement of the rotor shaft 5 at the upper electromagnet position. Calculate the displacements Dxm2 and Dim2.

The current control unit 443 performs proportional control, integral control, differential control, and the like based on the magnet position displacements Dxm1, Dim1, Dxm2, and Dim2, and generates a current setting signal related to the switching control of the excitation amplifier 43. For example, for one axis (x-axis direction) of the radial magnetic bearing 4A shown in FIG. 3, current setting signals for the magnetic bearing electromagnets 45m and 45p are generated, respectively. Then, a voltage equivalent signal for PWM control of the excitation amplifier 43 is generated based on the current deviation signal obtained by subtracting the current detection signal from the current sensor from the current setting signal.

The PWM calculation unit 444 generates a PWM control command based on the voltage equivalent signal from the current control unit 443. The amplifier drive unit 47 outputs a PWM gate drive signal to the excitation amplifier 43 based on the PWM control signal generated by the PWM calculation unit 444. The switching element provided in the exciting amplifier 43 is turned on and off by a PWM gate drive signal, and a voltage corresponding to the switching pattern is applied to the coil of the magnetic bearing electromagnet 45, whereby the exciting current of the magnetic bearing electromagnet 45 is controlled.

FIG. 6 is a diagram illustrating a conversion process in the displacement conversion unit 442. FIG. 6 shows the displacement in the x-axis direction, and is shown so that the ascent target position J in FIG. 3 coincides with the z-axis. Here, the upper displacement sensor position is zs1 and the lower displacement sensor position is zs2. Further, the upper electromagnet position is zm1 and the lower electromagnet position is zm2. The rotor shaft 5 is regarded as a rigid body, and the x-coordinates at the respective positions zs1, zs2, zm1 and zm2 of the central axis A of the rotor shaft 5 are xs1, xs2, xm1 and xm2.

Further, the position of the center of gravity of the pump rotor 3 (see FIG. 1) when the rotor shaft 5 is levitated to the ascent target position J is defined as CG. The distances from the center of gravity position CG to the displacement sensor positions zs1 and zs2 are L1 and L2, and the distances from the center of gravity position CG to the electromagnet positions zm1 and zm2 are l1 and l2.

In FIG. 6, since the ascent target position J is aligned with the z-axis, the x-coordinates xs1 and xs2 at the displacement sensor positions zs1 and zs2 represent the sensor position displacements Dxs1 and Dxs2 detected by the displacement sensor. Similarly, the x-coordinates xm1 and xm2 at the electromagnet positions zm1 and zm2 represent the magnet position displacements Dxm1 and Dxm2 at the electromagnet positions zm1 and zm2.

At this time, the magnet position displacements Dxm1 and Dxm2 are expressed by the following equations (7) and (8) using the sensor position displacements Dxs1 and Dxs2. In equations (7) and (8), the displacements Dxs1, Dxs2, Dxm1, and Dxm2 are described as functions of time t. Similarly, the magnet position displacements Dym1 (t) and Dym2 (t) in the y-axis direction are calculated by the following equations (9) (9) using the sensor position displacements Dys1 (t) and Dys2 (t) in the y-axis direction at the displacement sensor position. It is expressed as 10).
Dxm1 (t) = (l1-L1) / (L1 + L2) x {Dxs1 (t) -Dxs2 (t)} + Dxs1 (t) ... (7)
Dxm2 (t) = (-l2-L1) / (L1 + L2) x {Dxs1 (t) -Dxs2 (t)} + Dxs1 (t) ... (8)
Dim1 (t) = (l1-L1) / (L1 + L2) x {Dys1 (t) -Dys2 (t)} + Dys1 (t) ... (9)
Dym2 (t) = (-l2-L1) / (L1 + L2) x {Dys1 (t) -Dys2 (t)} + Dys1 (t) ... (10)

FIG. 7 is a diagram showing a frequency distribution of current amplitude and displacement amplitude when random noise is superimposed in the present embodiment. Line B1 shows the frequency distribution in the case of the conventional magnetic bearing and stay shown in FIG. 4B, and line B2 shows the frequency distribution in the case of the present embodiment. In the present embodiment, the displacements used for bearing control are the magnet position displacements Dxm1 (t), Dim1 (t), Dxm2 (t), and Dim2 (t) at the position of the magnetic bearing electromagnet 45, so that the frequency is low. Vibration due to the influence of noise in the region can be reduced.

(C1) As described above, in the first embodiment, as shown in FIG. 5, the magnetic bearing electromagnet 45 that magnetically levitates and supports the rotor shaft 5 in the radial direction and the displacement of the rotor shaft 5 in the radial direction are displaced. The displacement sensor 50 to be detected and the sensor position displacement Dxs1 (t), Dys1 (t), Dxs2 (t), Dys2 (t) detected by the displacement sensor 50 are converted into the magnet position displacement Dxm1 (at the position of the magnetic bearing electromagnet 45). Based on the displacement conversion unit 442 that converts to t), Dim1 (t), Dxm2 (t), Dim2 (t) and the magnet position displacement Dxm1 (t), Dim1 (t), Dxm2 (t), Dim2 (t). It is provided with a current control unit 443 that controls the exciting current of the magnetic bearing electromagnet 45 to control the levitation position of the rotor shaft 5 to a predetermined position.

By using the magnet position displacements Dxm1 (t), Dim1 (t), Dxm2 (t), and Dim2 (t) in this way, it is possible to suppress the occurrence of pump vibration due to the influence of noise superposition. For example, in a device such as an electron microscope that dislikes vibration, a vibration isolator for reducing vibration from the vacuum pump is often provided between the device and the vacuum pump, but in the present embodiment, there is no disturbance. Since the vibration can be reduced in the environment, it is possible to omit the vibration isolator. Further, by using the displacement of the magnet position, it is possible to reduce the time and the amount of work for returning the position of the rotor shaft 5 to the levitation target position J.

-Second embodiment-
FIG. 8 is a block diagram showing a schematic configuration of the magnetic bearing control system according to the second embodiment. In the configuration shown in FIG. 8, a signal selection unit 445 is further provided in the configuration of the block diagram shown in FIG. In the following, a part different from the configuration of FIG. 5 will be mainly described.

The sensor position displacements Dxs1, Dys1 (t), Dxs2, and Dys2 (t) output from the demodulation unit 441 are input to the displacement conversion unit 442 and also to the signal selection unit 445. The signal selection unit 445 includes sensor position displacements Dxs1 (t), Dys1 (t), Dxs2 (t), Dys2 (t) from the demodulation unit 441, and magnet position displacement Dxm1 (t) from the displacement conversion unit 442. Dim1 (t), Dxm2 (t), Dim2 (t) are input. When the magnitudes (that is, absolute values) of the sensor position displacements Dxs1, Dys1 (t), Dxs2, and Dys2 (t) of the signal selection unit 445 are equal to or less than a predetermined threshold value Dth, the magnet position displacement Dxm1 (t), Dim1 (t), Dxm2 (t), and Dim2 (t) are output to the current control unit 443. On the other hand, when the magnitude of any one of the sensor position displacements Dxs1 (t), Dys1 (t), Dxs2 (t), and Dys2 (t) exceeds the threshold value Dth, the signal selection unit 445 Sensor position displacement Dxs1 (t), Dys1 (t), Dxs2 (t), Dys2 (t) are output to the current control unit 443.

In an environment where the disturbance applied to the turbo molecular pump is very small, the pump vibration caused by the noise of the magnetic bearing system becomes a problem as described above. On the other hand, when a large disturbance such as an earthquake is applied to the turbo molecular pump, the pump vibration caused by the disturbance becomes a problem. The magnitude of the threshold value Dth in the signal selection unit 445 is, for example, the sensor position displacements Dxs1 (t), Dys1 (t), Dxs2 (t) when the rotor shaft 5 substantially contacts the emergency mechanical bearings 26a and 26b. , About 1/10 times that of Dys2 (t).

In this way, the displacement used for the current control is switched according to the magnitude of the displacement detected by the displacement sensor for the following reasons. When a large disturbance is applied and the rotor shaft 5 comes into contact with the mechanical bearings 26a and 26b, bending deformation of the rotor shaft 5 is induced. In a state where such bending deformation is large, the rotor shaft 5 is regarded as a rigid body and the magnet position displacements Dxm1 (t), Dim1 (t), Dxm2 (t), and Dim2 (t) are calculated as described above. There is a large error between the displacement and the actual displacement. Therefore, the stability of bearing control may be impaired. In particular, there is a risk that the levitation stability in a high frequency range, such as damping the natural frequency of bending, will be significantly impaired.

In such a case, from the viewpoint of robustness, it is more reliable to use the actually detected sensor position displacements Dxs1 (t), Dys1 (t), Dxs2 (t), and Dys2 (t). Therefore, in the second embodiment, if any one of the sensor position displacement Dxs1 (t), Dys1 (t), Dxs2 (t), and Dys2 (t) exceeds the threshold value Dth, the magnet position displacement Dxm1 ( Based on actually detected sensor displacements Dxs1 (t), Dys1 (t), Dxs2 (t), Dys2 (t), not t), Dym1 (t), Dxm2 (t), Dym2 (t) The exciting current is controlled.

(C2) As described above, in the second embodiment, as shown in FIG. 8, the values of the sensor position displacements Dxs1 (t), Dys1 (t), Dxs2 (t), and Dys2 (t) are within a predetermined range. When the magnitude of the value is included in the range of the threshold Dth or less, the magnetic bearing electromagnet 45 is excited based on the magnet position displacements Dxm1 (t), Dim1 (t), Dxm2 (t), and Dim2 (t). Control the current. Therefore, vibration caused by noise can be reduced in a low disturbance environment.

On the other hand, when the values of the sensor position displacements Dxs1 (t), Dys1 (t), Dxs2 (t), and Dys2 (t) are not included in the predetermined range (when the magnitude of the values exceeds the threshold Dth), The exciting current of the magnetic bearing electromagnet 45 is controlled based on the sensor position displacement Dxs1 (t), Dys1 (t), Dxs2 (t), and Dys2 (t). Therefore, it is possible to prevent the stability of the magnetic levitation control from being impaired when the magnitude of the displacement exceeds the threshold value Dth due to a large disturbance.

(C3) The current when the values of the sensor position displacements Dxs1 (t), Dys1 (t), Dxs2 (t), and Dys2 (t) are included in a predetermined range (the magnitude of the values is within the threshold Dth). The gain of the control unit 443 may be set to be smaller than the gain of the current control unit 443 when the value of the sensor position displacement is not included in the predetermined range.

As described with reference to FIG. 8, the current control unit 443 generates a current setting signal ist related to switching control of the excitation amplifier 43 based on the input displacement (sensor position displacement or magnet position displacement). Then, a voltage equivalent signal v for PWM control of the excitation amplifier 43 is generated based on the current deviation signal Δi obtained by subtracting the current detection signal from the current sensor from the current setting signal ist.

FIG. 9 is a transfer function block diagram relating to magnetic bearing control based on the current setting signal iset. The current control unit 443 is composed of PI control, that is, a proportional (P) element and an integral (I) element so as to reduce the steady-state deviation from the current setting signal ist as much as possible. 1 / (LS + R) is a transfer function for the magnetic bearing electromagnet 45, L is the inductance of the magnetic bearing electromagnet, and R is the resistance of the magnetic bearing electromagnet. The input unit is provided with a k gain that cancels 1 / k so that the input current setting signal ist corresponds to the output as it is.

The gain (G) of the entire current control unit is set to be larger than the gain (K) of the feedback line (G >> k). Usually, G is set as large as 1000 times or more of k, and in some cases up to about 100,000 times. As a result, it can be regarded as a linear amplifier having a constant gain in which the closed loop gain of the exciting amplifier becomes the inverse value (1 / k) of the feedback gain.

As described above, when the sensor position displacement is divided into a large region and a small region with the threshold Dth as a boundary, the gain G is set to a normal value in the region where the sensor position displacement is large, and in the region where the sensor position displacement is small. By reducing the gain G of the current control unit 443, the disturbance response is reduced in the region where the sensor position displacement is small, but the current fluctuation can be reduced, so that the vibration transmitted from the rotor shaft 5 to the pump body side is further reduced. be able to.

In the above-described embodiment, a turbo molecular pump having a turbo pump portion including a rotary blade 3a and a fixed blade 22 and a drag pump portion composed of a cylindrical portion 3b and a screw stator 24 will be described as an example. However, the present invention can be similarly applied to any vacuum pump having a configuration in which the pump rotor is magnetically levitated and supported by a magnetic bearing device. Further, the magnetic bearing device according to the present invention can be applied to the bearing mechanism of various devices, not limited to the vacuum pump.

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, 3 ... Pump rotor, 4A, 4B ... Radial magnetic bearing, 5 ... Rotor shaft, 33 ... Sensor circuit, 42 ... Motor, 43 ... Excitation amplifier, 44 ... Control unit, 45, 45m, 45p ... Magnetic bearing Electromagnet, 46,46m, 46p ... Current sensor, 50,50x1,50y1,50x2,50y2,50x1p, 50x1p, 51 ... Displacement sensor, 442 ... Displacement conversion unit, 443 ... Current control unit, 445 ... Signal selection unit, J ... Ascent target position

Claims (3)

A radial magnetic bearing electromagnet that magnetically levitates and supports the rotor shaft in the radial direction,
A displacement sensor that detects the radial displacement of the rotor shaft, and
A displacement conversion unit that converts the sensor position displacement detected by the displacement sensor into a magnet position displacement at the position of the radial magnetic bearing electromagnet.
A current control unit that controls the exciting current of the radial magnetic bearing electromagnet based on the displacement of the magnet position and controls the floating position of the rotor shaft to a predetermined position is provided .
When the value of the sensor position displacement is included in a range equal to or less than a predetermined threshold value, the exciting current of the radial magnetic bearing electromagnet is controlled based on the magnet position displacement.
Wherein when the value of the sensor position displacement is not included in the range, that controls the exciting current of the radial magnetic bearing electromagnet on the basis of the sensor position displacement, the magnetic bearing device. In the magnetic bearing device according to claim 1 ,
Wherein the gain of the current control unit is set smaller than the gain of the current controller when the value of the sensor position displacement is not included in the range in the case where the value of the sensor position displacement is included in the range, magnetic Bearing device. With the pump rotor
A motor that rotationally drives the pump rotor and
A vacuum pump comprising the magnetic bearing device according to claim 1 or 2 , which magnetically levitates and supports the rotor shaft of the pump rotor.

Images