vacuum pump

Abstract

Suppressing noise caused by motor switching. [Solution] A vacuum pump comprising a rotating body, a motor that rotates the rotating body, a radial electromagnet that generates a radial electromagnetic force on the rotating body, a displacement sensor having multiple sensor poles made of coils and measuring the displacement of the rotating body based on changes in the inductance of the coils, a power converter that converts an input DC voltage into an AC voltage to be applied to the motor, a motor driver that performs PWM control in cooperation with the power converter, an oscillator that outputs an AC voltage as a modulated input signal to the displacement sensor, and a demodulation circuit that demodulates the output signal from the displacement sensor using a synchronous detection method, wherein the carrier frequency in PWM control is four times the modulation frequency of the displacement sensor.

Description

Technical Field

[0001] The present invention relates to a vacuum pump.

Background Art

[0002] As background art in this technical field, there is Japanese Patent Application Laid-Open No. 2021-128133 (Patent Document 1). In this publication, in a vacuum pump, due to switching of a motor, noise is superimposed on the output of a displacement sensor used for levitation control of a magnetic bearing, and an improved method of a synchronous detection method for solving the problem of increased vibration and noise of the levitation control is disclosed (see FIG. 5). Further, in this Patent Document 1, it is pointed out that in the synchronous detection method, the noise component that is an odd multiple of the modulation frequency of the displacement sensor is large.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the configuration of Patent Document 1, the system design of synchronous detection becomes complicated. Therefore, the present invention focuses on various frequency constraints of a vacuum pump, and aims to provide a configuration that suppresses noise of a displacement sensor caused by switching of a motor after using a simple synchronous detection method.

Means for Solving the Problems

[0005] To solve the above problems, for example, the configuration described in the claims is adopted. This application includes a plurality of means for solving the above problems. For example, a vacuum pump, a rotating body, a motor that rotationally drives the rotating body, An electromagnet that generates an electromagnetic force on the rotating body, A displacement sensor having multiple sensor magnetic poles made of coils, which measures the displacement of the rotating body based on the change in the inductance of the coils, A power converter that converts an input DC voltage into an AC voltage to be applied to the motor, A motor driver that performs PWM control in cooperation with the aforementioned power converter, An oscillator that outputs an AC voltage as a modulated input signal to the displacement sensor, The system includes a demodulation circuit that demodulates the output signal from the displacement sensor using a synchronous detection method, The carrier frequency in the PWM control is four times the modulation frequency of the displacement sensor. [Effects of the Invention]

[0006] According to the present invention, noise in the displacement sensor caused by motor switching can be suppressed. Other issues, configurations, and effects not mentioned above will be clarified by the following description of the embodiments. [Brief explanation of the drawing]

[0007] [Figure 1] Figure 1 shows an example of a longitudinal cross-sectional view of a vacuum pump. [Figure 2] Figure 2 is an example of a circuit diagram for amplifier circuit 150. [Figure 3] Figure 3 shows an example of a timing chart illustrating control when the current command value is greater than the detected value. [Figure 4] Figure 4 shows an example of a timing chart illustrating control when the current command value is smaller than the detected value. [Figure 5] Figure 5A shows an example of the waveform of a PWM-controlled carrier signal. Figure 5B shows an example of the voltage output waveform due to the PWM control shown in Figure 5A. [Figure 6]FIG. 6A is a diagram showing the waveform of a sine wave voltage which is an output signal from an oscillator. FIG. 6B is an example of the time change of the displacement amount of the rotor shaft. FIG. 6C is a diagram showing the output signal of the displacement sensor. [Figure 7] FIG. 7 is a diagram showing the Fourier transform spectrum of the displacement sensor output signal. [Figure 8] FIG. 8A is an example of a circuit diagram for explaining synchronous detection by a demodulation circuit. FIG. 8B is a diagram showing the output waveform of the switch circuit. [Figure 9] FIG. 9 is a diagram showing the board diagram of a setting example of a band-pass filter. [Figure 10] FIG. 10 is the Fourier transform spectrum of the output signal of the switch circuit. [Figure 11] FIG. 11 is a diagram showing the board diagram of a setting example of a low-pass filter. [Figure 12] FIG. 12 is a diagram showing an example of the analysis result of the output signal of the demodulation circuit. [Figure 13] FIG. 13 is a diagram showing an example of the displacement amount detected by the displacement sensor in the first embodiment. [Figure 14] FIG. 14 is a diagram showing an example of the frequency spectrum of the time waveform of the displacement sensor signal. [Figure 15] FIG. 15 is an example of a longitudinal sectional view of the magnetic bearing device in the second embodiment. [Figure 16] FIG. 16 is a diagram showing an example of the spectrum of the demodulation circuit input signal from the radial displacement sensor in the second embodiment. [Figure 17] FIG. 17 is a diagram showing an example of the waveform of the PWM carrier signal in the third modification. [Figure 18] FIG. 18 is a diagram showing an example of the voltage utilization rate in the third embodiment.

MODE FOR CARRYING OUT THE INVENTION

[0008] <First Embodiment> Hereinafter, the vacuum pump 100 according to the first embodiment will be described with reference to the drawings. However, the installation mode of the vacuum pump 100 is not limited in any way. Also, for the same member, in one drawing, the member may be labeled, and in other drawings, the label may be omitted.

[0009] (1) Basic structure of the vacuum pump Hereinafter, preferred embodiments of the present invention will be described in detail with reference to FIG. 1. FIG. 1 is a diagram showing a schematic configuration example of the vacuum pump 100 according to an embodiment of the present invention, and is a diagram showing a cross-section in the axial direction along the rotation axis of the rotor shaft 113 in the vacuum pump 100. First, the vacuum pump 100 according to this embodiment will be described.

[0010] A longitudinal sectional view of this vacuum pump 100 is shown in FIG. 1. In FIG. 1, an intake port 101 is formed at the upper end of a cylindrical outer cylinder 127 of the vacuum pump 100. And inside the outer cylinder 127, a rotor 103 is provided with a plurality of rotating blades (rotating blades 102a, 102b, 102c...) which are turbine blades for sucking and exhausting gas, formed radially and in multiple stages on the circumferential part. A rotor shaft 113 is attached to the center of this rotor 103, and this rotor shaft 113 is levitated and position-controlled in the air by, for example, a magnetic bearing with five-axis control. The rotor 103 is generally made of a metal such as aluminum or an aluminum alloy.

[0011] Proximity to this upper radial electromagnet 104 and corresponding to each of the upper radial electromagnets 104, four upper radial displacement sensors 107 are provided. The upper radial displacement sensor 107 uses, for example, an inductance sensor having a conductive winding or an eddy current sensor, etc., and detects the position of the rotor shaft 113 based on the change in the inductance of this conductive winding that changes according to the position of the rotor shaft 113. The upper radial displacement sensor 107 has a plurality of sensor poles made of coils, and is configured to measure and detect the radial displacement of the rotor shaft 113, that is, the rotor 103 fixed thereto, based on the change in the inductance of the coil, and send it to the control device 200.

[0012] In this control device 200, for example, a compensation circuit having a PID adjustment function generates an excitation control command signal for the upper radial electromagnet 104 based on the position signal detected by the upper radial displacement sensor 107, and the amplifier circuit 150 (described later) shown in Figure 2 controls the excitation of the upper radial electromagnet 104 based on this excitation control command signal, thereby adjusting the upper radial position of the rotor shaft 113.

[0013] The rotor shaft 113 is made of a high-permeability material (such as iron or stainless steel) and is attracted by the magnetic force of the upper radial electromagnet 104. This adjustment is performed independently in the X-axis and Y-axis directions. The lower radial electromagnet 105 and lower radial displacement sensor 108 are arranged in the same way as the upper radial electromagnet 104 and upper radial displacement sensor 107, and adjust the lower radial position of the rotor shaft 113 in the same way as the upper radial position.

[0014] Furthermore, axial electromagnets 106A and 106B are positioned above and below a disc-shaped metal disk 111 located at the bottom of the rotor shaft 113. The metal disk 111 is made of a high-permeability material such as iron. An axial displacement sensor 109 is provided to detect the axial displacement of the rotor shaft 113, and its axial position signal is sent to the control device 200.

[0015] Then, in the control device 200, for example, a compensation circuit having a PID adjustment function generates excitation control command signals for the axial electromagnets 106A and 106B based on the axial position signal detected by the axial displacement sensor 109, and the amplifier circuit 150 excites the axial electromagnets 106A and 106B based on these excitation control command signals, so that the axial electromagnet 106A attracts the metal disk 111 upward with magnetic force, and the axial electromagnet 106B attracts the metal disk 111 downward, thereby adjusting the axial position of the rotor shaft 113.

[0016] Thus, the control device 200 appropriately adjusts the magnetic force exerted by the axial electromagnets 106A and 106B on the metal disk 111, causing the rotor shaft 113 to levitate axially and be held in contact with space. The amplifier circuit 150 that excites and controls the upper radial electromagnet 104, the lower radial electromagnet 105, and the axial electromagnets 106A and 106B will be described later.

[0017] On the other hand, the motor 121 is equipped with multiple magnetic poles arranged circumferentially around the rotor shaft 113. Each magnetic pole is controlled by the control device 200 to rotate the rotor shaft 113 via the electromagnetic force acting between it and the rotor shaft 113. The motor 121 also incorporates a rotational speed sensor, such as a Hall element, resolver, or encoder (not shown), and the rotational speed of the rotor shaft 113 is detected by the detection signal from this rotational speed sensor.

[0018] Furthermore, for example, a phase sensor (not shown) is attached near the lower radial displacement sensor 108 to detect the phase of rotation of the rotor shaft 113. The control device 200 uses both the detection signals from this phase sensor and the rotational speed sensor to detect the position of the magnetic pole.

[0019] Multiple fixed blades 123 (123a, 123b, 123c...) are arranged with a small gap between them and the rotating blades 102 (102a, 102b, 102c...). Each of the rotating blades 102 (102a, 102b, 102c...) is formed at a predetermined angle from a plane perpendicular to the axis of the rotor shaft 113 in order to transport exhaust gas molecules downward by collision. The fixed blades 123 (123a, 123b, 123c...) are made of metals such as aluminum, iron, stainless steel, copper, or alloys containing these metals as components.

[0020] Similarly, the fixed wing 123 is formed at a predetermined angle from a plane perpendicular to the axis of the rotor shaft 113, and is arranged alternately with the stages of the rotor blade 102 toward the inside of the outer cylinder 127. The outer edge of the fixed wing 123 is supported by being fitted between a plurality of stacked fixed wing spacers 125 (125a, 125b, 125c, etc.).

[0021] The fixed-wing spacer 125 is a ring-shaped member and is made of a metal such as aluminum, iron, stainless steel, or copper, or an alloy containing these metals as components. An outer cylinder 127 is fixed to the outer circumference of the fixed-wing spacer 125 with a small gap in between. A base portion 129 is provided at the bottom of the outer cylinder 127. An exhaust port 133 is formed in the base portion 129 and communicates with the outside. Exhaust gas that enters the intake port 101 from the chamber (vacuum chamber) side and is transferred to the base portion 129 is sent to the exhaust port 133.

[0022] Furthermore, depending on the application of the vacuum pump 100, a threaded spacer 131 is provided between the lower part of the fixed-blade spacer 125 and the base portion 129. The threaded spacer 131 is a cylindrical member made of a metal such as aluminum, copper, stainless steel, iron, or an alloy containing these metals, and has multiple helical screw grooves 131a engraved on its inner circumferential surface. The direction of the helix of the screw grooves 131a is such that when exhaust gas molecules move in the direction of rotation of the rotating body 103, these molecules are transported toward the exhaust port 133. A cylindrical portion 102d hangs down from the lowest part of the rotating body 103 following the rotor blades 102 (102a, 102b, 102c...). The outer circumferential surface of this cylindrical portion 102d is cylindrical and protrudes toward the inner circumferential surface of the threaded spacer 131, and is in close proximity to the inner circumferential surface of the threaded spacer 131 with a predetermined gap between them. The exhaust gas, which has been transferred to the screw groove 131a by the rotor blade 102 and the fixed blade 123, is guided through the screw groove 131a and sent to the base section 129.

[0023] The base portion 129 is a disc-shaped component that forms the base of the vacuum pump 100, and is generally made of a metal such as iron, aluminum, or stainless steel. The base portion 129 not only physically holds the vacuum pump 100 but also functions as a heat conduction path, so it is desirable to use a metal that is rigid and has high thermal conductivity, such as iron, aluminum, or copper.

[0024] In this configuration, when the rotor blade 102 is rotated by the motor 121 together with the rotor shaft 113, exhaust gas is drawn in from the chamber through the intake port 101 due to the action of the rotor blade 102 and the fixed blade 123. The rotational speed of the rotor blade 102 is usually 300Hz to 1500Hz, and the peripheral speed at the tip of the rotor blade 102 reaches 200m / s to 400m / s. The exhaust gas drawn in from the intake port 101 passes between the rotor blade 102 and the fixed blade 123 and is transferred to the base section 129. At this time, the temperature of the rotor blade 102 rises due to frictional heat generated when the exhaust gas comes into contact with the rotor blade 102 and heat conduction generated by the motor 121, but this heat is transferred to the fixed blade 123 side by radiation or conduction by gas molecules of the exhaust gas.

[0025] The fixed-wing spacers 125 are joined to each other at their outer circumference, and they transmit heat received by the fixed wing 123 from the rotor blade 102, as well as frictional heat generated when exhaust gases come into contact with the fixed wing 123, to the outside.

[0026] In the above description, the threaded spacer 131 is positioned on the outer circumference of the cylindrical portion 102d of the rotating body 103, and a threaded groove 131a is engraved on the inner surface of the threaded spacer 131. However, conversely, there are also cases where a threaded groove is engraved on the outer circumference of the cylindrical portion 102d, and a spacer having a cylindrical inner surface is positioned around it.

[0027] Furthermore, depending on the application of the vacuum pump 100, in order to prevent the gas drawn in from the intake port 101 from entering the electrical components, which consist of the upper radial electromagnet 104, upper radial displacement sensor 107, motor 121, lower radial electromagnet 105, lower radial displacement sensor 108, axial electromagnets 106A, 106B, and axial displacement sensor 109, the electrical components may be surrounded by a stator column 122, and the inside of this stator column 122 may be maintained at a predetermined pressure with purge gas.

[0028] In this case, piping (not shown) is provided in the base section 129, and purge gas is introduced through this piping. The introduced purge gas is sent to the exhaust port 133 through the gaps between the protective bearing 120 and the rotor shaft 113, between the rotor and stator of the motor 121, and between the stator column 122 and the inner cylindrical part of the rotor blade 102.

[0029] Here, the vacuum pump 100 requires model identification and control based on individually adjusted unique parameters (e.g., characteristics corresponding to the model). To store these control parameters, the vacuum pump 100 is equipped with an electronic circuit section 141 within its body. The electronic circuit section 141 consists of electronic components such as semiconductor memory such as EEP-ROM and semiconductor elements for accessing it, and a circuit board 143 for mounting them. This electronic circuit section 141 is housed, for example, below a rotational speed sensor (not shown) near the center of the base section 129 that constitutes the lower part of the vacuum pump 100, and is closed by an airtight bottom cover 145.

[0030] Incidentally, in the semiconductor manufacturing process, some process gases introduced into the chamber have the property of becoming solid when their pressure exceeds a predetermined value or their temperature falls below a predetermined value. Inside the vacuum pump 100, the exhaust gas pressure is lowest at the intake port 101 and highest at the exhaust port 133. If the process gas's pressure exceeds a predetermined value or its temperature falls below a predetermined value while it is being transferred from the intake port 101 to the exhaust port 133, the process gas will become solid and adhere to and accumulate inside the vacuum pump 100.

[0031] For example, if SiCl4 is used as the process gas in an Al etching apparatus, the low vacuum (760 [torr] ~ 10 -2 The vapor pressure curve shows that when the pressure is low (approximately 20°C), solid products (e.g., AlCl3) precipitate and adhere to the inside of the vacuum pump 100. As a result, when process gas precipitates accumulate inside the vacuum pump 100, these deposits narrow the pump flow path, causing a decrease in the performance of the vacuum pump 100. Furthermore, the aforementioned products were prone to solidifying and adhering in areas of high pressure, such as near the exhaust port 133 and near the threaded spacer 131.

[0032] Therefore, in order to solve this problem, conventional methods involve wrapping a heater (not shown) or an annular water-cooling pipe 149 around the outer circumference of the base portion 129, and embedding a temperature sensor (e.g., a thermistor) (not shown) in the base portion 129. Based on the signal from this temperature sensor, heating by the heater and cooling by the water-cooling pipe 149 are controlled (hereinafter referred to as TMS; Temperature Management System) to maintain the temperature of the base portion 129 at a constant high temperature (set temperature).

[0033] Next, we will describe an amplifier circuit 150 that energizes and controls the upper radial electromagnet 104, the lower radial electromagnet 105, and the axial electromagnets 106A and 106B of the vacuum pump 100 configured in this way. The circuit diagram of this amplifier circuit 150 is shown in Figure 8.

[0034] In Figure 8, the electromagnet winding 151, which constitutes the upper radial electromagnet 104, has one end connected to the positive terminal 171a of the power supply 171 via transistor 161, and the other end connected to the negative terminal 171b of the power supply 171 via current detection circuit 181 and transistor 162. Transistors 161 and 162 are so-called power MOSFETs, and have a structure in which a diode is connected between their source and drain.

[0035] In this configuration, transistor 161 has its diode cathode terminal 161a connected to the positive terminal 171a, and its anode terminal 161b connected to one end of the electromagnet winding 151. Transistor 162 has its diode cathode terminal 162a connected to the current detection circuit 181, and its anode terminal 162b connected to the negative terminal 171b.

[0036] On the other hand, the diode 165 for current regeneration has its cathode terminal 165a connected to one end of the electromagnet winding 151, and its anode terminal 165b connected to the negative terminal 171b. Similarly, the diode 166 for current regeneration has its cathode terminal 166a connected to the positive terminal 171a, and its anode terminal 166b connected to the other end of the electromagnet winding 151 via the current detection circuit 181. The current detection circuit 181 is composed of, for example, a Hall sensor type current sensor or an electrical resistance element.

[0037] The amplifier circuit 150 configured as described above corresponds to one electromagnet. Therefore, if the magnetic bearing is 5-axis controlled and there are a total of 10 electromagnets 104, 105, 106A, and 106B, a similar amplifier circuit 150 is configured for each electromagnet, and the 10 amplifier circuits 150 are connected in parallel to the power supply 171.

[0038] Furthermore, the amplifier control circuit 191 is configured, for example, by a digital signal processor (hereinafter referred to as the DSP section) of the control device 200 (not shown), and this amplifier control circuit 191 is configured to switch transistors 161 and 162 on and off.

[0039] The amplifier control circuit 191 compares the current value detected by the current detection circuit 181 (the signal reflecting this current value is called the current detection signal 191c) with a predetermined current command value. Based on this comparison, it determines the magnitude of the pulse width (pulse width time Tp1, Tp2) to be generated within the control cycle Ts, which is one period of PWM control. As a result, gate drive signals 191a and 191b with this pulse width are output from the amplifier control circuit 191 to the gate terminals of transistors 161 and 162.

[0040] Furthermore, when the rotating body 103 passes a resonance point during accelerated rotational speed operation, or when disturbances occur during constant-speed operation, it is necessary to control the position of the rotating body 103 with high speed and strong force. For this reason, a voltage of approximately 50V is used for the power supply 171 so that the current flowing through the electromagnet winding 151 can be rapidly increased (or decreased). In addition, a capacitor is usually connected between the positive electrode 171a and the negative electrode 171b of the power supply 171 to stabilize the power supply 171 (not shown).

[0041] In this configuration, when both transistors 161 and 162 are turned on, the current flowing through the electromagnet winding 151 (hereinafter referred to as the electromagnet current iL) increases, and when both are turned off, the electromagnet current iL decreases.

[0042] Furthermore, by turning one of transistors 161 and 162 on and the other off, a so-called flywheel current is maintained. By allowing this flywheel current to flow through the amplifier circuit 150, hysteresis loss in the amplifier circuit 150 can be reduced, and the overall power consumption of the circuit can be kept low. In addition, by controlling transistors 161 and 162 in this way, high-frequency noise such as harmonics generated in the vacuum pump 100 can be reduced. Moreover, by measuring this flywheel current with the current detection circuit 181, the electromagnet current iL flowing through the electromagnet winding 151 can be detected.

[0043] In other words, if the detected current value is smaller than the current command value, both transistors 161 and 162 are turned on only once during the control cycle Ts (e.g., 100 μs) for a duration corresponding to the pulse width time Tp1, as shown in Figure 9. Therefore, the electromagnet current iL during this period increases from the positive electrode 171a to the negative electrode 171b, towards the current value iLmax (not shown) that can flow through transistors 161 and 162.

[0044] On the other hand, if the detected current value is greater than the current command value, both transistors 161 and 162 are turned off only once during the control cycle Ts for a duration corresponding to the pulse width time Tp2, as shown in Figure 10. Therefore, the electromagnet current iL during this period decreases from the negative electrode 171b towards the positive electrode 171a, towards a regenerative current value iLmin (not shown) via diodes 165 and 166.

[0045] In either case, after the pulse width time Tp1 and Tp2 have elapsed, one of transistors 161 or 162 is turned on. Therefore, during this period, the flywheel current is maintained in the amplifier circuit 150.

[0046] (2) PWM control of motor 121 Next, the PWM control of the motor 121 will be described in detail. Figure 5A shows an example of the waveform of the PWM control carrier signal. The vacuum pump 100 further includes a power converter that converts the input DC voltage into an AC voltage to be applied to the motor 121, and a motor driver that performs PWM control in cooperation with the power converter.

[0047] As shown in Figure 5A, a triangular wave is used as the PWM control carrier signal. In this embodiment, the period of the PWM control carrier signal is 10 μs (frequency 100 kHz). Inside the motor driver, the command values ​​for each phase voltage are updated with each control cycle. In this embodiment, the control cycle is 40 μs (frequency 25 kHz).

[0048] Figure 5B shows the U-phase voltage output waveform under PWM control as shown in Figure 7A, when the DC link voltage is 140V (140V for the upper arm, 0V for the lower arm). As shown in Figure 5B, the U-phase voltage is +140V when the U-phase PWM command value is greater than the PWM carrier signal, and 0V when it is smaller. In this way, the wave of the PWM carrier period is repeated within the same control period.

[0049] (3) Regarding displacement sensor signals Figure 6A shows the waveform of a sinusoidal voltage, which is the output signal from an oscillator (not shown) in this embodiment (or the input signal or modulation signal of the displacement sensor). The frequency of this modulation signal (displacement sensor modulation frequency) must be sufficiently larger than the frequency band in which displacement is to be detected (e.g., 1500 Hz), for example, 25 kHz (period 40 μs). This sinusoidal voltage may include an offset component. Figure 6B shows an example of the time variation of the rotor shaft displacement 113. In this example, the displacement of the rotor shaft 113 is sinusoidal with a frequency of 1 kHz (period of 1000 μs).

[0050] Figure 6C shows the output signal of the displacement sensor under these conditions. The displacement sensor output signal is a carrier wave modulated by the displacement of the rotor shaft 113 by the displacement sensor input signal. The displacement sensor output signal is the product (or a value proportional to the product) of the displacement sensor input signal and the rotor displacement.

[0051] For example, in an inductance-type displacement sensor, this waveform is obtained by utilizing the fact that the inductance of the displacement sensor changes with the displacement of the rotor shaft 113 (the gap between the rotor shaft 113 and the displacement sensor). However, although not shown in the diagram, in reality, eddy currents inside the displacement sensor cause amplitude changes and phase lags that depend on the frequency of the displacement of the rotor shaft 113.

[0052] Figure 7 shows the Fourier transform spectrum of the displacement sensor output signal in Figure 6C. As shown in Figure 7, the 1 kHz component is almost absent in the Fourier transform spectrum of the displacement sensor output signal, and sidebands at 24 (=25-1) and 26 (=25+1) kHz appear. These sidebands represent modulated displacement information, which is demodulated by the subsequent demodulation circuit.

[0053] (4) Synchronous detection using demodulation circuits Figure 8A is an example of a circuit diagram illustrating synchronous detection using a demodulation circuit. Figure 8B shows the output waveform of a switch circuit. As shown in Figure 8A, the demodulation circuit of the present invention employs a synchronous detection method. The synchronous detection method has the characteristic of being able to easily reduce random noise and noise of a specific phase synchronized with the modulation signal. In the demodulation circuit, the carrier wave-like sensor output signals from the upper radial displacement sensor 107, the lower radial displacement sensor 108, and the axial displacement sensor 109 are first input to the bandpass filter BPF. That is, the displacement sensors in this invention include radial displacement sensors 107 and 108, and axial displacement sensor 109. The electromagnets in this invention include radial electromagnets 104 and 105 and axial electromagnets 106A and 106B.

[0054] A bandpass filter (BPF) removes signals from the input carrier signal in frequency bands far from the displacement sensor modulation frequency. This is because displacement information is contained in the sidebands of the displacement sensor modulation frequency, and frequency components far from the displacement sensor modulation frequency are noise. Ideally, the characteristics of the bandpass filter (BPF) should be such that the amplitude and phase of the sidebands of the displacement sensor modulation frequency, which represent displacement information, are hardly altered.

[0055] Figure 9 shows a Bode plot of an example bandpass filter (BPF) configuration. As shown in Figure 9, the bandpass filter (BPF) is designed to minimize the amplitude and phase changes within 1500 Hz before and after the displacement sensor modulation frequency of 25 kHz.

[0056] The switch circuit shown in Figure 8A receives the output signal of the bandpass filter (BPF) and its phase-inverted version as inputs. Additionally, a square wave synchronized with the displacement sensor modulation signal is generated and input to the switch circuit as a reference signal. The frequency of the reference signal is equal to the frequency of the displacement sensor modulation signal, which in this example is 25 kHz. The phase of the reference signal is determined considering the phase lag and other characteristics of the displacement sensor, and therefore is not necessarily in phase with the displacement sensor modulation signal. When the reference signal is High, the switch circuit outputs the sensor output signal without phase inversion. Conversely, when the reference signal is Low, the switch circuit outputs the sensor output signal with phase inversion.

[0057] As a result, as shown in Figure 8B, the output signal of the switch circuit can be obtained as a waveform obtained by folding the sensor output signal back every half period of the displacement sensor modulation signal. Figure 10 is the Fourier transform spectrum of the output signal of the switch circuit shown in Figure 8B. However, since the example in Figure 8B does not contain noise components, calculations including the effect of the bandpass filter (BPF) have not been performed. It can be seen that the 1000 Hz frequency component is dominant, indicating that the displacement information has been demodulated. However, the output signal of the switch circuit shows sidebands around even multiples of the displacement sensor modulation frequency, such as 49 (=2 × 25 - 1), 51 (=2 × 25 + 1), and 99 (=4 × 25 - 1) kHz.

[0058] The output signal of the switch circuit is input to a low-pass filter (LPF). Since the frequency of the displacement signal component is sufficiently smaller than the frequency of the sidebands around even multiples of the displacement sensor modulation frequency, this allows for the reduction of only the sidebands around even multiples of the displacement sensor modulation frequency, thereby extracting the displacement information. Figure 11 shows a Bode plot of an example low-pass filter (LPF) configuration. It is designed to minimize gain and phase changes up to approximately 1500 Hz.

[0059] The above demodulation method makes it possible to extract true displacement information from the carrier wave, which is the output signal of the displacement sensor. The output signal of this demodulated low-pass filter (LPF) is input to an AD converter (ADC) and used for controlling the magnetic bearing.

[0060] (5) For each frequency Next, as an example of this embodiment, a specific example in which the vacuum pump 100 is a turbomolecular pump will be described. In this embodiment, the frequencies of the vacuum pump 100 are as follows: • Rotation frequency: 500Hz • Displacement sensor modulation frequency: 25kHz • Magnetic bearing control frequency: 25kHz • Magnetic bearing carrier frequency: 25kHz • Motor control frequency: 25kHz • Carrier frequency of motor 121: 100kHz Each of these frequencies is explained below.

[0061] The rotational frequency is the rotational frequency of the rotor shaft 113. The rotational frequency is determined by the required exhaust performance of the vacuum pump 100. The rotational frequency is set to, for example, a maximum of approximately 1500 Hz.

[0062] The displacement sensor modulation frequency is the frequency of the displacement sensor modulation signal. As mentioned above, the displacement sensor modulation frequency must be sufficiently greater than the rotation frequency expressed in terms of mechanical angle. Furthermore, it is desirable that the displacement sensor modulation frequency be an integer of the magnetic bearing control frequency. This is because, in a suitable configuration, the displacement sensor modulation signal is controlled by the control device 200 that generates the control signal for the magnetic bearing.

[0063] In magnetic bearing control, the on / off time ratio of transistors 161 and 162 of the magnetic bearing control device 200 is determined at predetermined time intervals based on the detected displacement of the rotor shaft 113 and the electromagnet current. The reciprocal of this time interval is the magnetic bearing control frequency. In order to stably levitate the rotor shaft 113 using the magnetic bearing, the magnetic bearing control frequency must be sufficiently greater than the rotation frequency.

[0064] The magnetic bearing carrier frequency is the frequency of the PWM control used to switch the on / off state of transistors 161 and 162 of the magnetic bearing control device 200. Since the PWM control of the magnetic bearing is performed by the magnetic bearing controller, it is desirable that the magnetic bearing carrier frequency be an integer multiple of the magnetic bearing control frequency.

[0065] In motor control, the motor driver determines the on / off time ratio of the motor's power converter transistor at predetermined time intervals, based on the detected rotation angle of the rotor shaft 113 and the motor current. The reciprocal of this time interval is the motor control frequency. It is desirable that the motor control frequency be sufficiently greater than the rotation frequency expressed in electrical angle.

[0066] Furthermore, it is desirable for motor control and magnetic bearing control to be synchronized with each other. This is because simultaneous AD conversion and transistor switching can easily lead to noise superimposing on the AD conversion. However, by synchronizing motor control and magnetic bearing control, it becomes easier to implement switching prohibition times for each PWM control, thereby preventing the switching of each transistor during AD conversion and preventing noise superimposition on the AD conversion. For this reason, it is desirable for the motor control frequency to be an integer multiple or an integer fraction of the magnetic bearing control frequency.

[0067] The carrier frequency of motor 121 is the frequency of the PWM control used to switch the on / off state of the transistor in the power converter of motor 121. It is desirable that the carrier frequency of motor 121 be an integer multiple of the motor control frequency.

[0068] If the carrier frequency of motor 121 is made too high, the heat generated by the transistors and diodes that make up the power converter increases. Also, the ratio of the dead time and the switching prohibition time for AD conversion to the carrier period (the reciprocal of the carrier frequency) increases, so the voltage utilization rate of motor 121 decreases. On the other hand, if the carrier frequency of motor 121 is made too low, the current harmonics of motor 121 increase, and the iron loss and vibration of motor 121 increase.

[0069] Here, the output signal of the displacement sensor is superimposed with noise from the switching of the motor 121's PWM control. The dominant components of this noise are sideband components centered around the motor 121's carrier frequency and its integer multiples, separated by the rotation frequency expressed in electrical angles. For example, if the motor 121 has a carrier frequency of 25 kHz, a rotation speed of 1 kHz, and 2 poles, the main components of the noise are 24 (=25-1), 26 (=25+1), 49 (=2×25-1), 51 (=2×25+1) kHz, etc. Thus, the frequency of the noise from the motor 121 depends on the carrier frequency of the PWM control. In order to reduce the impact of the motor 121's noise while considering various constraints of equipment other than the motor 121, it is useful to change the motor 121's carrier frequency.

[0070] Figure 12 shows the analysis results of the output signal of the demodulation circuit (the input signal to the AD converter) when only noise of the same single frequency signal is assumed as the input signal to the demodulation circuit (with the carrier wave of the displacement sensor output signal set to 0). However, the displacement sensor modulation frequency in the analysis model is set to 25 kHz, and the characteristics of the bandpass filter (BPF) and low-pass filter (LPF) are as shown in Figures 9 and 11, respectively.

[0071] Furthermore, due to the characteristics of the demodulation method, the maximum frequency component of the output signal and the frequency of the input signal do not match, but the maximum frequency component of the output signal for each frequency of the input signal is plotted. For example, input signals of 24kHz and 26kHz result in an output signal of 1kHz. In addition, the analysis results for a very narrow bandwidth around odd multiples of the displacement sensor modulation frequency are greatly affected by the phase of the noise component, but this does not affect the overall nature of the present invention, so only one condition for the phase of the noise component is plotted.

[0072] As shown in Figure 12, there is a peak of approximately 5 kHz around the displacement sensor modulation frequency (25 kHz) and its odd multiples (75, 125, 175, ... kHz), with the output being higher compared to the areas before and after this peak. Furthermore, a deep dip can be observed directly above the displacement sensor modulation frequency (25 kHz) and its odd multiples (75, 125, 175, ... kHz) depending on the noise phase conditions. However, as mentioned above, the actual noise frequency components are shifted from these frequencies by the rotation frequency expressed in electrical angles, resulting in a larger output. In other words, synchronous detection is not very effective at reducing noise components that are slightly shifted from the displacement sensor modulation frequency and its odd multiples.

[0073] Furthermore, around 5 kHz near even multiples of the displacement sensor modulation frequency (50, 100, 150, 200, ... kHz), the output is lower compared to the surrounding frequencies. This indicates that synchronous detection significantly reduces noise around even multiples of the displacement sensor modulation frequency.

[0074] Furthermore, for noise above the displacement sensor modulation frequency (25 kHz), the output signal is approximately halved when the frequency increases by 50 kHz. This is because the bandpass filter (BPF) in the demodulation circuit reduces the high-frequency components before the switch circuit. Furthermore, since the frequency components of the noise input to the demodulation circuit and the frequency components appearing in the output signal do not necessarily coincide, the contribution of the low-pass filter (LPF) to noise around these frequencies is not always significant.

[0075] In conventional vacuum pumps, the carrier frequency of the motor 121 was often set to be the same as or about half the modulation frequency of the displacement sensor. This was because, in the structure of conventional vacuum pumps, the noise component of the motor 121 entering the demodulation circuit was small, and there was no need to further reduce this noise. However, when considering a structure that brings the displacement sensor closer to the conventional motor 121 in order to miniaturize the vacuum pump, this noise component became impossible to ignore.

[0076] As shown in Figure 12, setting the carrier frequency of motor 121 to an even multiple of the displacement sensor modulation frequency and increasing it leads to a reduction in noise components. On the other hand, increasing the carrier frequency of motor 121 leads to an increase in the heat generated by the transistors and diodes that make up the power converter.

[0077] Furthermore, increasing the carrier frequency of motor 121 leads to an increase in the ratio of dead time and switching prohibition time for AD conversion to the carrier period (reciprocal of carrier frequency), thus reducing the voltage utilization rate of motor 121. The switching prohibition time for AD conversion is, for example, 1 μs.

[0078] Inside a vacuum pump, dielectric breakdown is likely to occur at the vacuum connector terminals, based on Paschen's law. To prevent this, the DC link voltage of the motor 121 and magnetic bearing must be reduced, and these voltages are set to, for example, 140V. Thus, there is an upper limit to the DC link voltage of the vacuum pump's motor 121 and magnetic bearing, making it difficult to increase these voltages to compensate for the decrease in voltage utilization. As a result, the current required for motor 121 increases as the voltage utilization rate decreases. This increase in motor 121 current leads to increased heat generation in the transistors and diodes that make up the power converter, increased copper loss in motor 121, and the need for larger windings in motor 121.

[0079] (6) Effects of the present invention This invention focuses on the modulation and demodulation method of the displacement sensor of a vacuum pump and the carrier frequency of the motor 121, and derives optimal values ​​while considering various constraints related to these, thereby reducing noise caused by the motor 121 superimposed on the displacement sensor output. By setting the carrier frequency of the motor 121 to exactly four times the displacement sensor modulation frequency, which is a larger value than conventional methods, the noise caused by the motor 121 superimposed on the displacement sensor output is reduced while minimizing the deterioration of the voltage utilization rate of the motor 121. <Example 1> Next, we will explain the advantages of the motor 121 having a carrier frequency exactly four times that of the displacement sensor modulation frequency, based on an example where there is noise due to the difference frequency from four times.

[0080] The conditions in this embodiment are as follows: • Rotation frequency 400Hz • Displacement sensor modulation frequency: 25,000 Hz • Carrier frequency of motor 121: 100,033 Hz

[0081] Figure 13 shows the time waveform of the measured displacement output by the demodulation circuit in Example 1. Figure 14 shows the frequency spectrum of the time waveform in Figure 13. As shown in Figure 13, the time waveform of the displacement sensor signal shows a beat-like phenomenon in the displacement signal. As shown in Figure 14, in addition to the 400Hz rotation frequency signal, numerous sidebands of approximately 33Hz can be observed. This is thought to be due to the difference frequency between the displacement sensor modulation frequency (or an integer multiple thereof) and the motor 121 carrier frequency. For this reason, it is desirable for the displacement sensor modulation frequency and the motor 121 carrier frequency to be synchronized. In particular, if different hardware is used for the displacement sensor modulation signal and the motor 121 PWM control, it is advisable to use a synchronization signal to make one a master and the other a slave, or to prepare a separate master controller and make both slaves.

[0082] <Example 2> Next, we will explain why reducing motor noise superimposed on displacement sensors has become important due to changes in the structure of vacuum pumps resulting from miniaturization and other objectives.

[0083] First, the configuration of the magnetic bearing device 1 (5-axis controlled magnetic bearing) installed in the vacuum pump 100 in Example 2 will be described. Figure 15 is an example of a longitudinal cross-sectional view of the magnetic bearing device 1 in Example 2. As shown in Figure 15, the magnetic bearing device 1 is broadly composed of a shaft assembly 2 and a stator assembly 3, with the stator column 4 containing all but a portion of the rotor shaft 20 on the intake port side.

[0084] The shaft assembly 2 is composed of a rotor shaft 20, an upper radial displacement sensor target 21 fixed to the rotor shaft 20, a first spacer 22, an upper electromagnet target 23, a second spacer 24, a rotor 25, a third spacer 26, a lower radial displacement sensor target 29, a fourth spacer 28, a lower electromagnet target 27, and a retainer 30, all arranged in this order from the intake side (upper side) to the exhaust side (lower side).

[0085] The stator assembly 3 is configured by arranging the following components in this order from the intake port side to the exhaust port side: upper protective bearing 5, upper radial displacement sensor 6 (displacement sensor of the present invention), upper radial electromagnet 7 (electromagnet of the present invention), stator 8, lower radial displacement sensor 10 (displacement sensor of the present invention), lower radial electromagnet 9 (electromagnet of the present invention), lower protective bearing 11, upper axial electromagnet 12, axial spacer 13, armature disk 14, axial displacement sensor target 15, lower axial electromagnet 16, and axial displacement sensor 17. The following provides a detailed explanation of each of the above-mentioned configurations.

[0086] First, a motor 40 for rotating the rotor shaft 20 at high speed is provided in the middle of the rotor shaft 20 in the axial direction. The motor 40 comprises a stator 8 attached to the outer circumference of the rotor shaft 20 and a rotor 25 arranged to surround the stator 8 from the radial outside. The stator 32 is connected to a control device 200, and the rotation of the rotor shaft 20 is controlled by the control device 200.

[0087] The motor 40 is equipped with multiple magnetic poles arranged circumferentially around the rotor shaft 20. Each magnetic pole is controlled to rotate the rotor shaft 20 via an electromagnetic force acting between it and the rotor shaft 20. The motor 40 also incorporates a rotational speed sensor, such as a Hall element, resolver, or encoder (not shown), and the rotational speed of the rotor shaft 20 is detected by the detection signal from this rotational speed sensor.

[0088] The vacuum pump 100 further comprises a power conversion unit and a motor driver. The power conversion unit is a power converter that converts the input DC voltage into an AC voltage to be applied to the motor 40. The motor driver works in cooperation with the power converter to perform PWM control, thereby controlling the voltage applied to each magnetic pole of the motor 40.

[0089] Furthermore, a phase sensor (not shown) is attached near, for example, the upper radial displacement sensor 6 or the lower radial displacement sensor 10 to detect the phase of rotation of the rotor shaft 20. The control device 200 uses both the detection signals from this phase sensor and the rotational speed sensor to detect the position of the magnetic pole.

[0090] Furthermore, as a radial magnetic bearing device for non-contact support of the rotor shaft 20 in the radial direction, an upper radial electromagnet 7 and an upper electromagnet target 23 are provided on the intake side of the motor 40. The upper radial electromagnet 7 generates a radial electromagnetic force.

[0091] On the other hand, on the exhaust port side of the motor 40, a pair of lower radial electromagnets 9 and lower electromagnet targets 27 are provided. The lower radial electromagnets 9 generate a radial electromagnetic force. Each electromagnet target (23, 27) is fixed to the rotor shaft 20. The magnetic force of the electromagnets in this radial magnetic bearing device (electromagnet and electromagnet target) attracts the rotor shaft 20.

[0092] Note that Figure 15 describes the configuration of the magnetic bearing device 1 on the left side relative to the center line of the rotor shaft 20, but the other side has a similar configuration. That is, the magnetic bearing device 1 has four electromagnets arranged around the rotor shaft 20, separated by a predetermined clearance, so as to face each other at 90-degree intervals.

[0093] The upper radial displacement sensor 6 and upper radial displacement sensor target 21, and the lower radial displacement sensor 10 and lower radial displacement sensor target 29 are elements that detect radial displacement of the rotor shaft 20, and the displacement sensors have multiple sensor magnetic poles, for example, made of coils.

[0094] The coils of the upper radial displacement sensor 6 and the lower radial displacement sensor 10 are part of an oscillation circuit formed in a control device 200 (see Figure 1) installed outside the vacuum pump 100. A high-frequency current flows as the oscillation circuit oscillates, generating a high-frequency magnetic field on the rotor shaft 20. The upper radial displacement sensor 6 and the lower radial displacement sensor 10 measure the radial displacement of the rotor shaft 20 based on the change in the inductance of their respective coils.

[0095] In other words, for example, when the distance between the upper radial displacement sensor 6 and the upper radial displacement sensor target 21 changes, the oscillation amplitude in the oscillation circuit changes, making it possible to detect the displacement of the rotor shaft 20 on the upper side in the axial direction. Also, when the distance between the lower radial displacement sensor 10 and the lower radial displacement sensor target 29 changes, the oscillation amplitude in the oscillation circuit changes, making it possible to detect the displacement of the rotor shaft 20 on the lower side in the axial direction.

[0096] Normally, non-contact upper radial displacement sensors 6 use inductance-type or eddy current-type displacement sensors. However, in order to reduce variations in the output signal due to individual differences in the upper radial displacement sensor target 21 and its installation, an inductance-type displacement sensor is used in this case.

[0097] Thus, the magnetic bearing device 1 of this embodiment is a so-called five-axis controlled magnetic bearing device in which the rotor shaft 20 is held radially by a radial magnetic bearing device, and axially by an axial magnetic bearing device, and rotated around the axis.

[0098] The conditions in this embodiment are as follows: • Rotation frequency: 462Hz • Displacement sensor modulation frequency: 25kHz • Carrier frequency of motor 40: 100kHz

[0099] Figure 16 shows the frequency spectra of the measured values ​​of the demodulation circuit input signals from the upper radial displacement sensor 6 and the lower radial displacement sensor 10 in Example 2. In this figure, Xh is the demodulation circuit input signal spectrum of the upper radial displacement sensor 6, which is on the side furthest from the motor 40. Xb is the demodulation circuit input signal spectrum of the lower radial displacement sensor 10, which is on the side closer to the motor 40.

[0100] In Figure 16, the 25kHz ± 462Hz signal is the true displacement signal component, and the 100kHz ± 462 × n Hz (where n is an integer) signal is the noise component due to the switching of the motor 40. That is, it can be confirmed that the noise component at Xb, which is closer to the motor 40, is about 20dB (10 times) larger than the true displacement signal component. In this embodiment, the lower radial displacement sensor 10 is positioned in the axial direction between the motor 40 and the lower radial electromagnet 9, which is why the configuration of the present invention is particularly effective.

[0101] <Example 3> Next, we will explain the relationship between carrier frequency and voltage utilization rate. Here, we focus on the switching prohibition time for current detection AD conversion of the motor 121 in the vacuum pump 100 shown in Figure 1. In a typical design, for example, as shown in Figure 17, switching may be restricted for 0.5 μs before and after the trough of the triangular wave of the PWM carrier frequency (1 μs in total). We will consider the voltage utilization rate based on this.

[0102] Figure 18 is a plot of an example of the voltage utilization rate in Example 3. In Example 3, in order to reduce the influence of motor 121 noise on the displacement sensor, a configuration is adopted in which the carrier frequency of motor 121 is increased while being selected from among displacement sensor modulation frequency × an even number. As shown in Figure 18, in this design example, if the carrier frequency of motor 121 is four times the modulation frequency of the displacement sensor, a voltage utilization rate of up to 0.90 can be secured.

[0103] In a standard design for the vacuum pump 100, increasing the carrier frequency of the motor 121 while selecting from (displacement sensor modulation frequency × even number) significantly reduces the voltage utilization rate (in this example, it decreases by 0.05 for each increase). This is because the vacuum pump 100 requires a relatively high carrier frequency due to the high-speed rotation of the rotor shaft 113, and the motor 121 current and the switching prohibition time for magnetic bearing displacement detection AD conversion are relatively large in this application. Therefore, in the design of the vacuum pump 100, it is common practice to design the carrier frequency of the motor 121 to be as small as possible while maintaining current control of the motor 121.

[0104] The carrier frequency of motor 121 that can reduce the noise of motor 121 superimposed on the displacement sensor is an even multiple of the modulation frequency of the displacement sensor. Candidates around 4 times include 2 times or 6 times. However, if the carrier frequency of motor 121 is 2 times the modulation frequency of the displacement sensor, the amount of noise reduction is insufficient, especially in the miniaturized vacuum pump 100. On the other hand, if the carrier frequency of motor 121 is 6 times the modulation frequency of the displacement sensor, the voltage utilization rate decreases even more significantly (0.85 in this example), leading to increased heat generation in the vacuum pump 100, etc. Therefore, setting the carrier frequency of the motor 121 to four times the modulation frequency of the displacement sensor is considered optimal when taking into account the characteristics of the vacuum pump 100.

[0105] (7) Others The radial electromagnet and radial displacement sensor may be integrated into a single unit. That is, the radial electromagnet and radial displacement sensor may be formed as a single radial magnetic bearing unit by winding their respective coils around the same core. In this case, the multiple radial electromagnets are arranged in a circumferential direction. The coil of the radial displacement sensor is positioned between the multiple radial electromagnets in the circumferential direction.

[0106] It should be noted that the present invention is not limited to the embodiments described above, and various modifications are included. For example, the embodiments described above are described in detail for the purpose of explaining the present invention in an easy-to-understand manner, and are not necessarily limited to those having all the configurations described. Furthermore, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add and combine the configuration of one embodiment with the configuration of another embodiment. In addition, it is possible to add, delete, or replace parts of the configuration of each embodiment with other configurations. Furthermore, the embodiments described above disclose at least the configuration described in the claims. [Explanation of Symbols]

[0107] 100...Vacuum pump, 103...Rotating body, 104...Upper radial electromagnet, 105...Lower radial electromagnet, 107...Upper radial displacement sensor, 108...Lower radial displacement sensor, 109...Axial displacement sensor, 121...Motor, 200...Control device

Claims

1. It is a vacuum pump, A rotating body and A motor that rotates the aforementioned rotating body, An electromagnet that generates an electromagnetic force on the rotating body, A displacement sensor having multiple sensor magnetic poles made of coils, which measures the displacement of the rotating body based on the change in the inductance of the coils, A power converter that converts an input DC voltage into an AC voltage to be applied to the motor, A motor driver that performs PWM control in cooperation with the aforementioned power converter, An oscillator that outputs an AC voltage as a modulated input signal to the displacement sensor, The system includes a demodulation circuit that demodulates the output signal from the displacement sensor using a synchronous detection method, A vacuum pump in which the carrier frequency in the PWM control is four times the modulation frequency of the displacement sensor.

2. The motor, the radial electromagnet which generates a radial electromagnetic force on the rotating body, and the radial displacement sensor which measures the radial displacement of the rotating body are arranged in the axial direction along the rotation axis of the rotating body. The vacuum pump according to claim 1, wherein the radial displacement sensor is positioned in the axial direction between the motor and the radial electromagnet.

3. The system comprises multiple radial magnetic bearing units, each having an electromagnet that generates a radial electromagnetic force on the rotating body. In at least one of the radial magnetic bearing units, The vacuum pump according to claim 1, wherein the electromagnet and the displacement sensor are configured as a single unit.

Images