Factors to Consider when Selecting a Sensor
Sensors output electrical signals proportional (or at least we want them to be!) to the amplitude of the system parameter being measured. There are many physical phenomena that can be harnessed for generating the electrical signal; these include electrical resistance, capacitance and inductance to name a few. The choice of physical phenomena has a significant effect on the capabilities of the sensor, leading to each sensor having distinct advantages and disadvantages.
Clearly, if a sensor had no advantages compared to other sensors, it would not be commercially available. When selecting a sensor to purchase, we must decide which sensor is most appropriate by considering a wide variety of characteristics. The weight provided to each characteristic is unique to the application. The following list outlines common sensor characteristics that should be considered when selecting a sensor.
Cost: Often a deciding factor, given that there may be multiple sensors that meet or exceed our requirements.
Size and weight: Depending on the application, size and weight can be of paramount importance (e.g. in satellite systems) or negligible importance.
Mounting & electrical connections: Does the sensor include a mounting flange? What kind of electrical connector is required?
Output signal: The most common sensor output signals are 0-10 Vdc, 4-20 mA, binary and mV/V. The choice of output signal can effect your control systems susceptibility to background noise.
Excitation Voltage: Common input voltages include 5, 12, 24 and 48 V DC. Most sensors are capable of accepting a range of input voltages.
Power consumption: Most sensors consumer very little power and some consume none at all. Power consumption is important in battery powered applications such as remote sensing.
Measurement range: Sensors are designed to measure within set ranges of process variables. Sensors with large ranges suffer from reduced signal linearity and from increased error towards the lower bound of their range. Select a measurement range no larger than required.
Accuracy: The difference between the actual and measured value. It is composed of a number of factors including repeatability, hysteresis, and resolution.
Long term stability: Sensors experience drift over time which reduces their accuracy. A sensor with poor long term stability requires frequent re-calibration.
Lifetime: Many sensors undergo mechanical wear, limiting their lifetime. Lifetime is typically defined as the minimum number of cycles or hours of use before failure.
Resolution: Defined as the smallest change in the measured quantity that can be detected. Although most sensors are analog (i.e. hypothetically infinite resolution), practical resolution is limited by background noise and the DAC resolution of your process controller.
Sampling time: The minimum time between consecutive readings. It if often stated as sampling frequency, i.e. the number of readings per second.
Operating temperature: The temperature range within which the sensor can operate and achieve its stated accuracy.
Vibration/Shock: The maximum acceleration force that the sensor can survive.
Ingress protection: A standard code that rates the degree of protection provided by the sensors casing against infiltration of liquid and fine solid particles.
Contact/non-contact: The presence/absence of physical contact between the sensor and the environment being measured.
Operating pressure: The maximum gas or fluid pressure the sensor can withstand whilst achieving its stated accuracy.
Active/ Passive: An active sensor requires an external voltage to be applied (e.g. a strain gauge) whereas as a passive sensor does not (e.g. thermocouple).