Motor Shaft Angle Detection With Learned Pulse Switching Timing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing incremental and sensorless encoders face challenges in accurately determining the rotational position of a rotor magnet due to attachment errors of sensor magnets, leading to incorrect timing of U-phase, V-phase, and W-phase pulse signals.
Innovation Solution
An angle detection device and method that utilizes a sensor magnet on a rotor shaft with magnetic sensors to detect magnetic flux changes, storing a relational expression for calculating mechanical angles and outputting N-phase pulse signals with precise phase differences based on level switching angles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If three magnetic sensors are used to estimate mechanical angle without an absolute angle position sensor, then cost and size are reduced, but measurement precision deteriorates due to attachment errors of sensor magnets
Solution Approach 1:
The system performs preliminary learning processing before normal operation to establish the relationship between sensor output values and actual mechanical angles. During this learning phase, the motor is rotated through various positions and the arithmetic device stores the correspondence data in a storage device. This preliminary action compensates for attachment errors by creating a calibration map that corrects for misalignment between sensor magnet positions and sensor detection points, thereby maintaining measurement precision while using the simpler three-sensor configuration
Solution Approach 2:
The system changes the parameter representation from direct sensor output values to corrected mechanical angle values through learning processing. By storing and applying correction data that maps sensor outputs to actual positions, the system transforms the imprecise raw sensor data into accurate position information, effectively compensating for attachment errors without adding hardware complexity
2Adaptability or versatility
If sensor magnet is attached to rotation shaft, then sensorless motor operation is enabled, but attachment errors cause deviation from ideal zero-cross position leading to incorrect pulse signal timing
Solution Approach 1:
The system implements feedback through learning processing where the actual mechanical angle (from incremental encoder) is compared with the estimated angle (from three magnetic sensors). The difference or correction data is stored and used to adjust future angle calculations. This feedback mechanism continuously compensates for attachment errors, ensuring that pulse signal timing remains accurate despite variations in sensor magnet attachment positions
Solution Approach 2:
Before normal motor operation, the system performs preliminary learning processing to establish the relationship between sensor outputs and actual positions. This preliminary calibration creates a correction map that accounts for attachment errors, enabling the sensorless operation to achieve manufacturing precision comparable to systems with absolute position sensors
3Loss of information
If U-phase, V-phase, and W-phase pulse signals are output based on sensor signals, then rotational position indication is provided, but attachment errors cause incorrect level switching timing
Solution Approach 1:
The system uses feedback from the learning processing to correct the timing of level switching in pulse signals. By comparing the estimated mechanical angle from three magnetic sensors with the actual angle from the incremental encoder during learning, the system stores correction data that is applied during normal operation. This ensures that U-phase, V-phase, and W-phase pulse signals switch levels at the correct timing corresponding to the actual rotor magnet position, compensating for sensor magnet attachment errors
Solution Approach 2:
The system transforms the raw sensor output parameters into corrected parameters that account for attachment errors. By applying the learned relationship between sensor outputs and actual positions, the system adjusts the timing parameters of pulse signal level switching to match the ideal timing that would exist without attachment errors, thereby maintaining accurate rotational position information
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Accurately indicates the rotational position of the rotor magnet with high precision by synchronizing the level switching of pulse signals with the calculated mechanical angle, overcoming attachment errors.
Implementation Method 1
M magnetic sensors that detect a magnetic flux change due to rotation of the sensor magnet
Data Source
AI summary
An angle detection device includes a sensor magnet on a shaft of an N-phase motor having a rotor magnet, M magnetic sensors detecting a magnetic flux change due to rotation of the sensor magnet, a storage device storing an expression representing a relationship between an output value of the M magnetic sensors and a mechanical angle of the shaft corresponding to a rotational position of the rotor magnet as a level switching angle, and a processing device calculating the mechanical angle based on the output value and the expression, and outputting an N-phase pulse signal having a phase difference of 360 degrees divided by N in terms of electrical angle based on the calculated mechanical angle and the level switching angle. The processing device switches a level of a pulse signal of any one phase of the N-phase pulse signal when the calculated mechanical angle matches the level switching angle.