BLDC Motor Back EMF Zero Crossing Detection via Interpolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the zero crossing point of a phase winding in brushless DC motors face challenges such as noise interference and unreliable rotor position estimation, especially at low motor speeds, due to the conflict between maintaining a short undriven period for torque and noise reduction versus ensuring accurate detection of zero crossings.
Innovation Solution
The method involves driving the phase winding with a suitable waveform, providing an undriven period for sampling the back EMF signal, and interpolating these samples to determine the zero crossing point, allowing for accurate estimation even if the zero crossing occurs outside the undriven period, thus enabling short undriven periods and reduced noise while maintaining accurate motor position calculation across varying speeds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If the undriven period is made short to increase torque and reduce noise, then torque and noise performance improve, but the reliability of zero crossing detection deteriorates
Solution Approach 1:
The system performs preliminary sampling of back EMF signals at multiple points during the undriven period before the expected zero crossing occurs. By collecting sample data in advance and using interpolation to predict the zero crossing point, the system ensures accurate detection even when the undriven period is short, thus resolving the contradiction between maintaining short undriven periods for torque and ensuring reliable zero crossing detection.
2Measurement precision
If the undriven period is extended to ensure accurate zero crossing detection, then detection accuracy improves, but torque output and noise performance deteriorate
Solution Approach 1:
The system takes multiple samples of the back EMF signal during the undriven period and uses interpolation calculations to determine the zero crossing point. This preliminary sampling approach allows accurate detection without extending the undriven period, thereby maintaining short undriven periods that reduce noise and improve torque while still achieving high detection accuracy.
3Reliability
If conventional back EMF monitoring is used continuously, then rotor position monitoring is reliable, but device complexity and energy consumption increase
Solution Approach 1:
Instead of continuous monitoring, the system implements periodic sampling of the back EMF signal only during brief undriven periods in each phase cycle. By taking multiple discrete samples and using interpolation to calculate the zero crossing point, the system achieves reliable rotor position monitoring with reduced complexity and lower energy consumption compared to continuous monitoring approaches.
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
This approach enhances torque and reduces noise while providing reliable rotor position estimation at low motor speeds, enabling efficient operation without the need for constant back EMF monitoring, and allows for accurate zero crossing detection even if it occurs outside the undriven period.
Implementation Method 1
sampling the back EMF signal in the phase winding on two or more occasions during the undriven period
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
In a three phase BLDC motor the rotor position is monitored by detecting the zero crossing points of the induced back EMF signals BEMF_U, BEMF_V, BEMF_W in the phase windings U, V, W. As they are illustrated, the back EMF signals are substantially sinusoidal but they may in other situations be substantially trapezoidal. The three back EMF signals are 120° out of phase with each other. In order to accurately monitor the back EMF in a phase winding, the driving waveform for each phase U, V, W includes an undriven period P close to the expected zero crossing point. The period P can be a preset part of the driving waveform or can be an interruption of the normal driving waveform in response to suitable interrupt signals. In order to determine the zero crossing points of each back EMF signal, in the method of the present invention, two (or more) samples of the back EMF are taken during the undriven period P and used to interpolate the back EMF signal to determine the zero crossing point.