Much has been written about the promise of predictive analytics and how IoT data can improve operational efficiency, reduce downtime, and save money for the enterprise. In contrast, little is written about the sensors gathering the data that is fed into the predictive analytics engine. Panduit’s white paper, “E.S.P. for IoT Platforms,” discusses the characteristics to consider when deploying measurement sensors and how to determine the importance of specifications depending on sensor type and deployment location.
Sensor Types
There are three types of sensors: indicators, counters, and measurement.
- Indicators are relatively straight forward – they are either on or off. They show when something has occurred, for example, when someone has opened and accessed a panel.
- Counters can keep a running tally of a series of events. An example is a tachometer that counts the number of revolutions of a shaft or axle. Both indicators and counters are examples of digital sensors. They monitor and report discrete events. Relatively speaking, they are simple sensors.
- Measurement sensors are more sophisticated. They report on the amount of a physical entity, such as weight, or on an environmental attribute, such as temperature. Rather than reporting discrete events, they report where one is on a continuous scale.
Sensor Characteristics
When choosing sensors for your IoT platform, there are five characteristics you should consider.
- Accuracy
Accuracy is the ability of a sensor to provide a true measurement of whatever the sensor is monitoring. There is an uncertainty with the measurement, usually represented as a percentage of full scale. - Repeatability
Repeatability is the ability of a sensor to provide a constant output when there is a constant input, when acquiring a new sample. - Linearity
Linearity is a measure of how well the sensor’s response curve approaches a straight line. - Sensitivity
A sensor’s sensitivity is the amount the input to the sensor must change to detect any change in the output. - Environmental Impact
Changes in the environment can impact the performance and accuracy of a sensor. For example, some sensors are particularly sensitive to temperature and humidity.
When selecting a sensor, you should also determine which attributes are important to your application. In a benign environment, the environmental impact on the sensor’s performance may not be that important, whereas it may be a consideration if the application is outdoors.
The tradeoff you need to make when selecting a measurement sensor is the level of specificity you require for that attribute versus cost. For example, a temperature sensor monitoring a pizza oven does not need to be as accurate as one monitoring a pharmaceutical process. A temperature sensor with an accuracy of ±0.01°C will be more expensive than one with an accuracy of ±1°C.
To learn more about why sensors are important for your IoT platform, download Panduit’s
“E.S.P. for IoT Platforms” white paper – or subscribe to our blog to access all the papers in our IoT “101” white paper series.
Thanks for pointing out one of the commonly overlooked attributes – physical protection of the sensor. I’ve seen many users install sensors and find them damaged by water or dust infiltration. When selecting sensors make sure they meet the needs of the physical environment. An IP-20 rating is great for a data center. In a food processing plant, one may need an IP-66 rating. If the sensors will be exposed to caustic liquids or gas look for Nema 4X.