ECU Analog Input Calibration for Low-Precision AD Converters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing analog-digital converters (AD converters) face limitations in achieving high digital conversion precision due to linear errors and noise malfunctions, leading to temporary and critical defects in digital conversion values, especially when using inexpensive and low-precision AD converters.
Innovation Solution
An electronic control unit is designed with a multiplexer to select analog input signals from multiple points, applying them to an AD converter, and utilizing a microprocessor with added control programs for collinear approximate coefficient calculation, interpolation correction, averaging, and band checking to accurately calculate correction parameters, ensuring high-precision digital conversion data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If inexpensive and low-precision AD converters are used, then cost is reduced, but digital conversion precision deteriorates
Solution Approach 1:
The patent applies preliminary calibration action by measuring the actual conversion characteristics of the AD converter before use and storing correction data. The calibration is performed in advance using known reference voltages, and the correction parameters are stored for later application during normal operation, thereby improving precision without changing the hardware
Solution Approach 2:
The patent changes the parameters of the conversion process by introducing correction coefficients that modify the raw digital output. The correction parameters (slope and intercept) are calculated based on measured calibration data and applied to transform the inaccurate conversion results into precise values, effectively changing the conversion characteristic parameters
2Measurement precision
If digital calibration is performed to improve precision, then measurement precision is improved, but calculation errors may occur due to noise malfunction
Solution Approach 1:
The patent implements feedback by continuously monitoring the calibration process and using the measured actual conversion characteristics to adjust the correction parameters. The system feeds back the calibration results to refine the correction coefficients, ensuring that the calibration accurately reflects the actual converter behavior while compensating for noise and malfunctions
Solution Approach 2:
The patent applies beforehand cushioning by performing multiple calibration measurements and using statistical methods to determine the most reliable correction parameters. The calibration process includes preliminary checks and validation to cushion against the impact of noise malfunctions before final correction parameters are established
3Measurement precision
If correction parameters are calculated using reference analog signals, then digital conversion precision is improved, but the process time increases
Solution Approach 1:
The patent applies partial action by selecting only the essential calibration points needed to define the conversion characteristic (typically two points for linear correction). Rather than calibrating every possible input value, the method uses minimal necessary measurements to establish correction parameters that can then be applied across the entire conversion range, reducing calibration time while maintaining precision
Data Source
AI summary
An analog input signal obtained from an analog sensor group 104A and first and second calibration voltages obtained by high-precision voltage-dividing resistors are successively selected by a multiplexer, digitally converted through an AD converter and then input to a microprocessor. The microprocessor calculates a collinear approximate coefficient based on the first and second calibration voltages in cooperation with a program memory, and corrects the digital conversion value to the analog input signal by using the approximate coefficient, thereby correcting a linear error of the conversion characteristic of the AD converter. In the calculation of the approximate coefficient, upper and lower limit check is executed on measurement values and calculation coefficients, and also plural calculation results are averaged to enhance the precision.