In IO-Link applications repeatability, accuracy, and linearity are all specifications used to describe sensors.
But not all specifications are equal indicators of real-world performance; let's break it down.
Repeatability is the measurement of how reliably a sensor can duplicate the same measurement in the same conditions. Repeatability is calculated by having a sensor detect the same target multiple times under ideal settings.
Accuracy is the difference between the actual distance to a target and the sensor’s measured distance, and is important when using the measurements directly from the sensor via IO-Link.
But while highest accuracy may seem like the best, it’s not the most valuable specification when it comes to performance; if the application contains a known distance to the target, the difference between actual and measured values can be accounted for and calibrated out.
In these cases, another specification becomes more relevant to real-world performance: Linearity.
Linearity is the maximum deviation between an ideal straight-line measurement and the actual measurement. It indicates how closely a sensor’s analog output can approximate a straight line when measuring across the sensor’s range.
Linearity is the most valuable specification in known IO-Link applications because the more linear the sensor, the more easily any consistent inaccuracy can be calculated for and removed.