Displacement temperature drift detection and suppression method based on Hall sensor

Abstract

The invention provides a displacement temperature drift detection and suppression method based on a Hall sensor. The technical problem that when a temperature drift problem occurs in an eddy current sensor, the detection precision changes, and consequently an impeller works eccentrically is solved. Whether the eddy current sensor has a temperature drift problem or not is judged by comparing the signal amplitude difference between the Hall sensors which are symmetrically mounted relative to the physical center, and the impeller is guided to be finely adjusted towards the correct physical center position when the temperature drift problem occurs. And after the adjustment is completed, recording data of the eddy current sensor as a new physical center value, and correcting the temperature excursion of the eddy current sensor. According to the method, no extra hardware needs to be added, whether the temperature excursion problem exists in the eddy current sensor can be detected only through simple signal amplitude comparison of the Hall sensor, temperature excursion correction of the eddy current sensor is achieved, the method is not affected by radial hydraulic force, and the method is suitable for the pump liquid working scene of the magnetic suspension pump and has good application prospects. And the stability and the reliability of the system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ultra-clean magnetic levitation pumps, and particularly to a method for detecting and suppressing displacement temperature drift based on a Hall sensor. Background Art

[0002] Magnetic levitation pumps are a newly developed ultra-clean liquid pumping technology in recent years. Through non-contact electromagnetic bearing technology, the impeller is suspended in the pump chamber and rotates at high speed to pump liquid. It has the advantages of extremely low outlet pulsation, no mechanical friction, and almost no particle shedding. At the same time, it uses an ultra-pure fluororesin pump head, resulting in extremely low release of metal and ion pollutants, and shows great advantages in the application of various high-end semiconductor equipment in the 12-inch advanced process. It has become a core component commonly used in semiconductor machines such as single-wafer cleaning, electroplating / chemical plating, and chemical mechanical polishing (SDS system).

[0003] In a magnetic levitation pump, controlling the rotor at the central position is a necessary condition for the rotor to rotate stably at high speed for a long time. Eddy current displacement sensors can achieve non-contact displacement detection, and have the advantages of fast measurement response speed, wide measurement range, and compact structure. Therefore, in practice, the position of the rotor is usually detected by an eddy current displacement sensor. However, there is no mature industry detection standard for eddy current coils in the production process at present, and they are all wound by manufacturers themselves, so the coil consistency cannot be guaranteed. Therefore, when the operating environment changes, especially when affected by temperature, the performance of the eddy current sensor will fluctuate, and the detection accuracy will change, and this performance fluctuation cannot be predicted, which may lead to eccentric operation of the impeller, resulting in an increase in the power consumption of the whole machine; in severe cases, it will cause high-speed scraping between the impeller and the pump head, resulting in a large amount of particle shedding, polluting the semiconductor process production line and equipment. Therefore, a solution is needed to be able to detect in time when the eddy current sensor has a temperature drift problem and correct the data of the eddy current sensor.

[0004] The currently disclosed technical solutions mainly determine whether the motor rotor is truly at the central position by detecting the suspension current. When the suspension current is the smallest, the position where the rotor is located is the central position of the rotor. This method is used to adjust the rotor to the central position, and the value of the eddy current sensor is observed to judge whether there is a temperature drift problem. If so, calibration is performed. There are mainly two adjustment strategies, namely the perturbation observation method and the four-quadrant recursive search method. The perturbation observation method is to actively apply a step perturbation of any polarity, observe the changes in the d-axis and q-axis currents, and perform displacement compensation with the goal of reducing the absolute values of the d-axis and q-axis currents. The four-quadrant recursive search method is to define a cost function. When the cost function becomes larger, the stepping direction is switched according to the rule of rotating 90° clockwise until 360° switching is completed. After reducing the stepping length, the above steps are repeated until the stepping length is less than the termination stepping value.

[0005] The premise assumption of the above solution is that the suspension currents of the d and q axes are minimized when the impeller is in the central position. However, in the application of a magnetic levitation pump, when the pump is actually operating with liquid load, the impeller will be subject to hydraulic forces in the radial direction. Therefore, the position where the d and q axis currents are minimized is not the physical center position of the magnetic levitation pump volute. If the above solution is used to detect and correct the temperature drift problem of the eddy current sensor, it may lead to using the wrong physical center value as the basis for calibrating the eddy current sensor. In severe cases, it may cause high-speed rubbing between the impeller and the pump head. Summary of the Invention

[0006] To solve the technical problem that when the eddy current sensor has a temperature drift problem, the detection accuracy changes, resulting in eccentric operation of the impeller, a displacement temperature drift detection and suppression method based on Hall sensors is proposed. By comparing the signal amplitude differences between Hall sensors symmetrically installed relative to the physical center, it is determined whether there is a temperature drift problem with the eddy current sensor, and when a temperature drift problem occurs, the impeller is guided to make a fine adjustment to the correct physical center position. After the adjustment is completed, the data of the eddy current sensor is recorded as the new physical center value to achieve the correction of the temperature drift of the eddy current sensor.

[0007] The present invention uses the following device for detection: The device includes two pairs of Hall sensors and two pairs of eddy current sensors. The two pairs of Hall sensors and the two pairs of eddy current sensors are alternately distributed on the inner edge of the volute of the magnetic levitation pump. The adjacent two Hall sensors are spaced 90°, and the adjacent two eddy current sensors are spaced 90°. The connection lines of each pair of Hall sensors and each pair of eddy current sensors pass through the physical center of the volute and are symmetrically installed relative to the physical center of the volute. Taking the connection line of one pair of Hall sensors as the x-axis and the connection line of the other pair of Hall sensors as the y-axis. When the impeller is at the physical center, the physical distances of each pair of Hall sensors from the impeller are the same, and the magnetic field intensities they sense are the same. Therefore, the signal amplitudes received by the two opposite Hall sensors are the same. When the eddy current sensor has a temperature drift problem, at this time, the eddy current sensor will have a detection deviation, resulting in the upper computer controlling the impeller at the wrong physical center, making the signal amplitudes of the two opposite Hall sensors different. Therefore, the temperature drift and other problems of the eddy current sensor can be detected and corrected by the amplitude conditions of the Hall sensors.

[0008] The specific correction steps are as follows: Detect the signal amplitudes of the two pairs of Hall sensors. If the signal amplitude differences of the pair of Hall sensors on the x-axis and the signal amplitude differences of the pair of Hall sensors on the y-axis are both less than the set threshold, it can be determined that there is no temperature drift problem with the eddy current sensor; if the signal amplitude difference of any pair of Hall sensors on the x-axis and on the y-axis is greater than the set threshold, it can be determined that there is a temperature drift problem with the eddy current sensor, and the following steps are used for correction: Step 1. Set the initial step size step_start, the moving step size step, the termination step size step_end, the step size decay rate step_ratio (0 < step_ratio < 1), and the amplitude difference threshold amperr; Step 2. Detect the signal amplitudes of a pair of Hall sensors located on the x-axis and compare their magnitudes: If the difference between the two is less than or equal to amperr, stop the adjustment in the x-direction; if the difference between the two is greater than amperr, move a distance of the initial step size step_start in the direction of the Hall sensor with the smaller signal amplitude; Repeat the above steps until the magnitude relationship between the two changes, and then update the step size step to step = step * step_ratio. If the updated step is less than step_end, stop the adjustment in the x-direction; Step 3. Detect the signal amplitudes of a pair of Hall sensors located on the y-axis and compare their magnitudes: If the difference between the two is less than or equal to amperr, stop the adjustment in the y-direction; if the difference between the two is greater than amperr, move a distance of the initial step size step_start in the direction of the Hall sensor with the smaller signal amplitude; Repeat the above steps until the magnitude relationship between the two changes, and then update the step size step to step = step * step_ratio. If the updated step is less than step_end, stop the adjustment in the y-direction; Step 4. After completing the above adjustments, record the data of the eddy current sensor and observe whether there are any deviations in the data. Based on this, it can be judged whether there is a temperature drift problem with the eddy current sensor. If there are deviations in the data, use the newly recorded data as the physical center value detected by the eddy current sensor to correct the temperature drift of the eddy current sensor.

[0009] Using the method described in the present invention does not require additional hardware. By simply comparing the signal amplitudes of the Hall sensors, it can be detected whether there is a temperature drift problem with the eddy current sensor, and the temperature drift correction of the eddy current sensor can be achieved. Moreover, it is not affected by radial hydraulic forces and is applicable to the pump liquid working scenario of the magnetic levitation pump, improving the stability and reliability of the system. Description of the Drawings

[0010] Figure 1 It is a schematic diagram of the temperature drift problem of the eddy current sensor in the magnetic levitation pump in the embodiment. Detailed Embodiment

[0011] The present invention will be further described below with reference to the accompanying drawings.

[0012] As Figure 1As shown in the figure, two pairs of eddy current sensors and two pairs of Hall sensors are uniformly arranged at the inner edge 1 of the volute of the magnetic levitation pump with the physical center 2 of the volute as the origin, and are evenly cross-spaced.

[0013] The first eddy current sensor 5_1 and the third eddy current sensor 5_3 are symmetrically installed relative to the physical center 2 of the volute, and the second eddy current sensor 5_2 and the fourth eddy current sensor 5_4 are symmetrically installed relative to the physical center 2 of the volute; the first Hall sensor 6_1 and the third Hall sensor 6_3 are symmetrically installed relative to the physical center 2 of the volute, and the second Hall sensor 6_2 and the fourth Hall sensor 6_4 are symmetrically installed relative to the physical center 2 of the volute; the line connecting the second Hall sensor 6_2 and the fourth Hall sensor 6_4 is taken as the x-axis, and the line connecting the first Hall sensor 6_1 and the third Hall sensor 6_3 is taken as the y-axis. When the physical center 4 of the impeller 3 is at the physical center 2 of the volute, the physical distances between the first Hall sensor 6_1 and the third Hall sensor 6_3, the second Hall sensor 6_2 and the fourth Hall sensor 6_4, and the impeller 3 are the same, and the magnetic field intensities they sense are the same. Therefore, the signal amplitudes of the first Hall sensor 6_1 and the third Hall sensor 6_3 are the same, and the signal amplitudes of the second Hall sensor 6_2 and the fourth Hall sensor 6_4 are also the same.

[0014] When the eddy current sensor has a temperature drift problem, the eddy current sensor will have a detection deviation at this time, resulting in the control algorithm controlling the impeller at the wrong physical center (the physical center 4 of the impeller 3 is not at the physical center 2 of the volute), so that the signal amplitudes of the two relative Hall sensors are not the same.

[0015] Since the Hall sensor itself belongs to a commercial IC, its performance such as temperature drift has been strictly controlled during factory production. Inside the magnetic levitation motor, the temperature effects on each Hall sensor are the same, so the influence of temperature on its signal amplitude is basically the same; therefore, the influence of temperature on the Hall sensor does not need to be considered.

[0016] When the amplitude of the third Hall sensor 6_3 is greater than the amplitude of the first Hall sensor 6_1, and the amplitude of the fourth Hall sensor 6_4 is greater than the amplitude of the second Hall sensor 6_2, the temperature drift problem of the eddy current sensor can be detected and corrected by the amplitude conditions of the Hall sensors. The specific operation steps are as follows: Step 1. Set the initial step size step_start, the moving step size step, the termination step size step_end, and the step size attenuation rate step_ratio (0 <step_ratio<1）以及幅值差阈值amperr；Step 2. Detect the signal amplitudes of the second Hall sensor 6_2 and the fourth Hall sensor 6_4 located on the x-axis, denoted as V6_2 and V6_4 respectively, and compare their magnitudes: If |V6_2 - V6_4| ≤ amperr, stop the adjustment in the x direction; if |V6_2 - V6_4| > amperr, move a distance of an initial step size step_start in the direction of the Hall sensor with the smaller signal amplitude, that is, when V6_2 <v6_4，则将叶轮往第二霍尔传感器6_2方向移动一个移动步长step，初始移动步长为step_start，若v6_4<v6_2，则将叶轮往第四霍尔传感器6_4方向移动一个移动步长step，初始移动步长为step_start；Repeat the above steps until the size relationship between the two changes, and then update the step size step to step = step * step_ratio. If the updated step is less than step_end, stop adjusting in the x direction; Step 3. Detect the signal amplitudes of the first Hall sensor 6_1 and the third Hall sensor 6_3 located on the y-axis, and compare their magnitudes: If |V6_1 - V6_3| ≤ amperr, stop adjusting in the y direction; if |V6_1 - V6_3| > amperr, move a distance of one initial step size step_start in the direction of the Hall sensor with the smaller signal amplitude, that is, when V6_1 <v6_3，则将叶轮往第一霍尔传感器6_1方向移动一个移动步长step，初始移动步长为step_start，若v6_3<v6_1，则将叶轮往第三霍尔传感器6_3方向移动一个移动步长step，初始移动步长为step_start；Repeat the above steps until the size relationship between the two changes, and then update the step size step to step = step * step_ratio. If the updated step is less than step_end, stop the adjustment in the y direction; Step 4. After completing the above adjustment, record the data of the eddy current sensor and observe whether there is any deviation in the data. Based on this, it can be judged whether there is a temperature drift problem with the eddy current sensor. If there is a deviation in the data, use the newly recorded data as the physical center value detected by the eddy current sensor to correct the temperature drift of the eddy current sensor.

Claims

1. A method for detecting and suppressing displacement temperature drift based on Hall sensors, characterized in that: Two pairs of Hall sensors and two pairs of eddy current sensors are used to detect and suppress the temperature drift of the eddy current sensors. The two pairs of Hall sensors and the two pairs of eddy current sensors are alternately distributed on the inner edge of the volute of the magnetic levitation pump. The adjacent two Hall sensors are spaced 90°, and the adjacent two eddy current sensors are spaced 90°. The connection lines of each pair of Hall sensors and each pair of eddy current sensors pass through the physical center of the volute and are symmetrically installed relative to the physical center of the volute. The connection line of one pair of Hall sensors is used as the x-axis, and the connection line of the other pair of Hall sensors is used as the y-axis; Detect the signal amplitudes of the two pairs of Hall sensors, and based on the signal amplitude differences of the two pairs of Hall sensors, determine whether there is a temperature drift problem with the eddy current sensors, and when a temperature drift problem occurs, guide the impeller to make fine adjustments to the correct physical center position. After the adjustment is completed, record the data of the eddy current sensors as the new physical center value to achieve the correction of the temperature drift of the eddy current sensors.

2. The method for detecting and suppressing displacement temperature drift based on a Hall sensor according to claim 1, characterized in that: By comparing the signal amplitude differences between the Hall sensors symmetrically installed relative to the physical center: If the signal amplitude differences of a pair of Hall sensors located on the x-axis and the signal amplitude differences of a pair of Hall sensors located on the y-axis are both less than the set threshold, it can be determined that there is no temperature drift problem with the eddy current sensors; If the signal amplitude difference of any pair of Hall sensors located on the x-axis and on the y-axis is greater than the set threshold, it can be determined that there is a temperature drift problem with the eddy current sensors.

3. The method for detecting and suppressing displacement temperature drift based on Hall sensors according to claim 1, characterized in that: The following steps are used to correct the temperature drift of the eddy current sensors: Step 1. Set the initial step size step_start, the moving step size step, the termination step size step_end, the step size attenuation rate step_ratio (0 < step_ratio < 1), and the amplitude difference threshold amperr; Step 2. Detect the signal amplitudes of a pair of Hall sensors located on the x-axis and compare their magnitudes: If the difference between the two is less than or equal to amperr, stop the adjustment in the x direction; If the difference between the two is greater than amperr, move a distance of an initial step size step_start in the direction of the Hall sensor with the smaller signal amplitude; Repeat the above steps until the magnitude relationship between the two changes, and then update the step size step to step = step * step_ratio. If the updated step is less than step_end, stop the adjustment in the x direction; Step 3. Detect the signal amplitudes of a pair of Hall sensors located on the y-axis and compare their magnitudes: If the difference between the two is less than or equal to amperr, stop the adjustment in the y direction; If the difference between the two is greater than amperr, move a distance of an initial step size step_start in the direction of the Hall sensor with the smaller signal amplitude; Repeat the above steps until the size relationship between the two changes, and then update the step length step to step = step *step_ratio. If the updated step is less than step_end, stop adjusting in the y direction. Step 4. After completing the above adjustments, record the data of the eddy current sensor and observe whether there is any deviation in the data. This can be used to determine whether the eddy current sensor has a temperature drift problem. If there is a deviation in the data, use the newly recorded data as the physical center value detected by the eddy current sensor to correct the temperature drift of the eddy current sensor.

Images