1. How do inductive proximity sensors work?
Inductive proximity sensors are non-contact sensors that detect the proximity of a metal (ferrous or non-ferrous) target to the sensors face through the interaction of the target with an electromagnetic field generated by the sensor. They output either a Boolean signal (i.e. target detected or not detected) or an analog signal proportional to the targets distance from the sensor. The inductive proximity sensor consists of a copper coil wound around an iron core (forming an electromagnet), an LC oscillator circuit and a current meter. The coil is excited by an AC voltage generated by the oscillator circuit (a capacitor placed in series with the coil, forming an LC circuit). The resulting AC current passing through the coil generates an alternating electromagnetic field. When a metal target comes within range of the sensor, it acts as a load, resisting the alternating electromagnetic field and thereby reducing the current flowing through the coil. The reduction in current is measured by the current meter located inside the sensor.
Inductive proximity sensors emit either analog or Boolean output signals. Analog output type inductive proximity sensors convert the reduction in current to a 0-10 V or 4-20 mA output which is (ideally) a linear function of the distance between the target and sensor. Boolean output type inductive proximity sensors use a comparator to convert the Hall voltage to a logic 0 or logic 1 output depending on whether a threshold value of current has been crossed. The comparator used in proximity sensors is called a Schmitt trigger, which differs from a regular comparator in that it has built in hysteresis. The practical effect of the hysteresis is illustrated below; as the target moves to within 5 mm of the sensor, the sensor is triggered high. However, the sensor will only be triggered low once the target moves to at least 6 mm away. This prevents an oscillation of the sensors output state about a particular value.
Inductive proximity sensors are sensitive to the type of metal from which the target is made. Factory calibration is carried out using a 1 mm thick plate of mild steel. Using a target made of a different metal will affect the sensing range. Furthermore, it will cause a fixed percentage error in the output signal of an analog sensor. To account for different types of metals, an analog type sensors output voltage or Boolean type sensors threshold distance are multiplied by a reduction factor (table below). By definition mild steel has a reduction factor of 1. A brass target (reduction factor of 0.5) at a distance of 5 mm from the sensor will induce the same sensor reading as a mild steel target at a distance of 10 mm.
2. General characteristics of inductive proximity sensors
Inductive proximity sensors are low to medium cost sensors designed for measuring the proximity of metal (ferrous or non-ferrous) objects at distances up to 40 mm (up to 75 mm is available in non-standard housings). They are available with either analog or Boolean output signals. The sensors have poor output linearity and are sensitive to the type of metal being detected. As such, analog output models are not recommended for applications where an accuracy of greater than 10% of full scale is required. Inductive proximity sensors are non-contact and solid state meaning that they are easy to assemble, do not experience mechanical wear and have an extremely long lifetime. They typically operate at sampling rates of 0.2-1 kHz, though some may have sampling rates as high as 5 kHz. As well as having excellent ingress protection, their readings are not affected by contamination of the sensor surface or target with non-conductive dirt, debris or liquid.
3. Input and output signals
Inductive proximity sensors operate on a 4.5-36 Vdc supply voltage (a minority of sensors operate on an AC supply voltage). The AC coil voltage is generated using an internal LC oscillator circuit. Analog output sensors convert the measured coil current to a 0-10 V or 4-20 mA output signal. Boolean output sensors convert the measured coil current to either 0 V for a logic 0 or the supply voltage (less the voltage drop across the transistor) for a logic 1, in accordance to whether a threshold current has been crossed. If the sensors output is normally open (NO) then movement of the target to within sensing range will cause the sensors output to switch from logic 0 to logic 1. If the sensors output is normally open (NC) then movement of the target to within sensing range will cause the sensors output to switch from logic 1 to logic 0. Most Boolean output sensors have either a NO or a NC output, with the exception or 4 wire sensors which have both a NO and a NC output.
4. Applications of inductive proximity sensors
Inductive proximity sensors are used in a wide range of position measurement applications where accurate position measurement is NOT required. This includes environments with significant dirt and oil contamination. Typical applications include use as non-contact limit switches in machines and to count metallic objects on a conveyor belt. They can be also used to count rotations of a shaft or gear up to 5000 counts per seconds (i.e. 5,000 Hz) though most models operate at only 200-1,000 Hz.
5. Typical specification
|Measurement range||2-75 mm*|
|Sampling rate||0.2-1 kHz (up to 5 kHz available)|
|Velocity||Limited only by sampling rate|
|Hysteresis||10-20% (Boolean models only)|
|Lifetime||Limited only by sampling rate|
|Ambient temperature||-25 to 70°C typical (up to 250 °C available)|
|Supply voltage||4.5-36 Vdc (10-250 Vac available)|
|Output signal||Boolean or analog|
|Vibration resistance||25 g|
|Shock resistance||50-100 g|
|Ingress protection||IP67 typical (IP69 available)|
|Active / passive||Active|
|Contact / non-contact||Non-contact|
*Assuming a mild steel target. Furthermore, the maximum sensing distance for sensors housed in a standard threaded cylinder is 40 mm. To ensure reliable triggering of Boolean output sensors, aim for the actual separation distance to be at least 10% smaller than the stated triggering distance.
6. Purchasing tips
- Flush or non-flush housings: Proximity sensors are available in either flush or non-flush housings. Flush housings extend the entire length of the sensor, enabling the sensor face to be mounted ‘flush’ to a wall (hence the name). Non-flush housings do not extend the full length of the sensor. The flush housing also acts as an electromagnetic shield and so flush housed sensors are shielded and non-flush housed sensors are not. The shielding partially protects the sensors coil from external electromagnetic fields but also reduces the sensors measurement range and field of view.
- Weld immune: Some inductive proximity sensors are described by their manufacturers as ‘weld immune’. These sensors are suitable for welding applications (and therefore other harsh environments) because they have increased immunity to external electromagnetic fields. Furthermore, they usually include stainless steel housings and ceramic coated sensor faces for protection from welding slag.
- Supply voltage: Most inductive proximity sensors operate on a DC supply voltage. Sensors which operate on an AC supply voltage typically have sampling frequencies of only 10 Hz. In comparison, those operating on DC supply voltage may have sample rates exceeding 1 kHz.
- NO/NC: Boolean inductive proximity sensors are available with either normally open (NO) outputs, normally closed (NC) outputs, or both.
- Number of wires: Analog inductive proximity sensors may have 2 or 3 wires and Boolean inductive proximity sensors may have 2, 3 or 4 wires. Sensors with only 2 wires often lack reverse polarity and short circuit protection. Boolean output sensors with 3 wires have either a normally open (NO) or normally closed (NC) output. Boolean output sensors with 4 wires have both a NO and a NC output.
- Adjustable threshold: Many Boolean output sensors enable the user to adjust the threshold measurement distance at which the sensor output switches states.
- PNP/NPN: The transistor used to switch output states (in response to the Schmitt trigger) can be of either PNP or NPN type (also known as current sourcing and current sinking types respectively). When interfacing the sensor to a PLC, it is vital that the sensor is of the opposite type to the PLC port e.g. a current sourcing sensor will only work with a current sinking PLC port.
- Housing: Although most inductive proximity sensors are housed in small externally threaded cylinders, there are alternative housing designs available. The alternative designs are often used to accommodate larger diameter inductance coils, enabling larger measurement ranges.
- Sample rate: For applications where the sample rate of the inductive proximity sensor is insufficient, it is possible to use an eddy current sensor instead. These operate on the same principal but with an air core instead of an iron core inside the copper coil. This reduces the resistance to the alternating electromagnetic field, allowing the eddy current sensor to operate at sampling frequencies in the MHz range.
7. Advantages of inductive proximity sensors
Inductive proximity sensors:
- Are of low cost and suitable for use in harsh environments (with the exception of high EMI). There are models available which are suitable for operating temperatures as high as 250 °C.
- Can detect all types of metals. Furthermore, they cannot sense non-metals, which is an advantage in some applications. For example, an inductive sensor can sense a metal target through a plastic machine guard.
- Accuracy is not reduced by contamination of the sensor face or target with non-conductive dirt, debris or liquid.
8. Disadvantages of inductive proximity sensors
Inductive proximity sensors:
- Have poor signal linearity (as do Hall effect and capacitive sensors) and are sensitive to changes in temperature. Although this is not a problem for Boolean output sensors, analog output sensors are not recommended for applications where an accuracy or more than ±10% of full scale range is required.
- Are sensitive to the type of metal being detected. Although this is corrected by the reduction factor, it still causes a loss of accuracy. The loss of accuracy is due to slight variations in reduction factor across different alloys of a metal.
- Are sensitive to the size of the target. Using a target smaller than the face of the sensor will reduce the sensing range.
- Are affected by external electromagnetic fields. Check the sensors EMI rating to determine if this will be an issue.
- Are not able to sense non-metals, limiting their range of applications
- Have a sampling rate of 0.2-1 kHz, not high enough for all applications.
9. Application tips
- Accuracy: The output linearity of inductive proximity sensors is very low at either extreme of their measurement range. As such, when using an analog output inductive sensor, aim to avoid operating within the first or last 10% of its measurement range. Many inductive sensors saturate as the target is brought to within 1 mm of the senor face. As such, an analog output sensor with an advertised measurement range of 0-5 mm, may only have an actual range of 1-5 mm.
- Installation: DO NOT countersink a non-flush sensor into a metal casing as this could result in a false signal as the sensor detects the metal casing.
- Multiple sensors: Multiple Boolean output sensors can be connected in series or parallel to a single process controller input port to achieve logic functions. Two sensors connected in series act as an AND gate. Two sensors connected in parallel act as an OR gate. The circuits used to implement parallel and series connections can be found here
- Leakage current: The leakage current (i.e. the residual current flowing when the output is at logic 0) of a 2 wire Boolean output sensor may result in a sufficiently high voltage to cause a reset error. This can be avoided by placing a bleeder resistor in parallel with the controller.