Performance Metric
Degree of proportionality between the output of a sensor and the input across several discrete increments of the full measurement scale. Technicians evaluate multi point linearity by comparing the actual sensor readings against a series of known reference values. This measurement determines the maximum deviation from a straight line drawn between the zero and full scale points.
Calibration Curve
Increasing the number of test points provides a more detailed map of the sensor response than a simple two point check. To assess multi point linearity, the instrument is subjected to loads at ten percent intervals of its rated capacity. The resulting data allow for the calculation of the non linearity error as a percentage of the full scale output.
This method captures deviations that a sparse data set would miss.
Signal Correction
Digital signal processors use the data from these tests to apply mathematical corrections to the raw output. If the multi point linearity shows a consistent curve, a polynomial equation can linearize the response in real time. Accurate modeling reduces uncertainty.
Hysteresis Interaction
Differences between the upward and downward measurement cycles sometimes obscure the underlying linearity. When measuring multi point linearity, it is common to perform both an ascending and a descending run. Hysteresis often limits the achievable linearity.