The sensor is a transducer that converts a physical property into an electrical signal. The physical property can be Weight, Temperature, Pressure, Percentage Composition, Force, Electric or Magnetic or Electromagnetic, Position and Orientation, etc.
The sensors are classified as active sensors and passive sensors based on their working principle. The active sensors use an external or self-generated signal to measure. For example, RADAR emits a series of intense radio waves for a short time and waits for the radio waves or signal to return and calculate the distance of the distant object using the signal flight time. The passive sensors work by changing their electrical properties like resistance or capacitance based on the physical property. For example, an LDR changes its resistance based on the intensity of light.
Sensors must have the following significant properties to define the quality of a sensor:
Every sensor has a range in which they work with an acceptable error. If the input is not in range, then the output is unpredictable.
The signal level varies for the same input over a long period; this is called as drift. The drift will cause an error in the measured value. The drift may result from aging of the sensor or temperature variance.
Sensitivity is defined as the change in output per unit change in input of the property being measured. The sensitivity of the sensor may be constant or linear for the entire range of sensor or vary exponentially if the sensor is a non-linear sensor
Selectivity is the ability of the sensor to measure a target property in the presence of other properties. For example, if an oxygen sensor does not react to other gasses like CO2 then it has good selectivity.
The resolution of a sensor is the minimum change in the target property that can produce a detectable change in output. For example, consider a temperature sensor with a resolution of 1C; this temperature sensor cannot produce a different output for 0.1C change in input.
Response and Recovery Time
The response time is the time taken by the sensor for its output to reach 95% of its final value when it is exposed to a target material. The Recovery Time is defined conversely.
If the sensitivity of the sensor is constant for the range, then it is called as linearity of the sensor. The linear sensors are easier to use while the non-linear sensors require complex mathematical equations to measure the physical property.
The hysteresis is the characteristic of a sensor by which the sensor produces a different set of outputs if the data is recorded in different directions (increasing input or decreasing input). The hysteresis can be observed in the following figure
If a meaningful measurement is to be made, it is necessary to tune the output of the sensor with accurately known input.
The full-scale output is the difference between the output for maximum input and the output for minimum input. Based on this, the ADC’s reference voltages have to be selected properly.
The precision of a sensor is its ability to produce same output when repeatedly measured for the same input. The precision is determined using statistical analysis standard deviation.
The accuracy of a sensor defines how close the output is to the real value. The accuracy defines the maximum error the sensor may produce.