Self-Calibrating Phase Estimation for Encoder Detectors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing encoder systems with widely separated detectors and non-uniform performance characteristics face errors due to mechanical misalignment and varying signal strengths, which degrade the accuracy of phase estimation for absolute angular position sensing.
Innovation Solution
A self-calibrating phase estimation method and apparatus that uses a primary correlation phase processor and pre- and post-quadrature calibration subsystems to adaptively compensate for magnitude, phase, and offset errors in quasi-sinusoidal signals from multiple detectors, generating accurate phase angle estimates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If detectors are widely separated in encoder systems, then the system can be more flexible in mechanical arrangement, but errors due to mechanical misalignment and varying signal strengths increase
Solution Approach 1:
The system performs preliminary calibration by measuring the actual phase shifts and magnitude variations of each detector signal relative to an ideal reference. These calibration parameters are stored and used to correct subsequent measurements, compensating for the effects of wide detector separation and mechanical misalignment before phase estimation occurs.
Solution Approach 2:
The calibration process determines actual phase shift parameters and magnitude parameters for each detector, then applies these parameter corrections to the raw detector signals. This transforms the signals from having varying, error-prone parameters to having corrected, uniform parameters suitable for accurate phase estimation.
2Adaptability or versatility
If detectors have non-uniform performance characteristics, then individual detector variations can be accommodated, but the accuracy of phase estimation degrades
Solution Approach 1:
The system measures the actual phase shift and magnitude parameters for each detector and applies corrective transformations to equalize their performance characteristics. This converts detectors with non-uniform parameters into effectively uniform detectors, enabling accurate phase estimation despite initial variations.
Solution Approach 2:
The calibration process provides feedback about each detector's actual performance parameters, which are then used to adjust the signal processing to compensate for deviations from ideal behavior. This closed-loop approach ensures that detector variations are continuously corrected.
3Device complexity
If simple processing is applied to perfect sinusoidal signals, then the processing complexity is low, but real encoder signals are not perfect and require advanced processing
Solution Approach 1:
The system performs preliminary calibration to determine correction parameters for each detector, then applies these corrections to transform real, imperfect sinusoidal signals into idealized signals. This preliminary processing step enables the use of simpler phase estimation algorithms while maintaining high accuracy.
Solution Approach 2:
The calibration parameters act as an intermediary between the imperfect real signals and the ideal processing algorithms. By introducing these correction factors, the system bridges the gap between realistic signal conditions and simplified processing methods.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Phase estimation apparatus processes sensor signals from sensors to estimate a phase of a periodically varying state of an object, such as position of a moving object. A phase estimation processor applies a first correlation calculation to simultaneously collected samples of the sensor signals to generate first quadrature values, where the first correlation calculation employs variable calculation values, and applies a phase calculation to the first quadrature values to generate the phase estimation. A pre-quadrature calibration circuit applies respective second correlation calculations to respective sequences of samples of the sensor signals individually to generate second quadrature values for each of the sensor signals, and applies phase and/or magnitude calculations to the sets of second quadrature values to generate the variable calculation values for the first correlation calculation, thereby compensate for the error component and improve accuracy of the estimated phase.