A proximity sensor detects that an object is nearby without ever touching it. Where a mechanical limit switch relies on physical contact and moving parts, a proximity sensor senses through a field, which makes it faster, more durable, and well suited to dirty or high-cycle environments. This guide explains what a proximity sensor is, the difference between inductive and capacitive types, and what PNP and NPN outputs mean when wiring one.
Proximity Sensor in one line: A proximity sensor is a non-contact device that detects the presence or absence of a nearby object and switches an output accordingly, without physically touching the object. Inductive types sense metal by disrupting an electromagnetic field, while capacitive types sense metals and non-metals by a change in capacitance. Because there are no moving parts to wear, proximity sensors are used for reliable position and speed detection in demanding conditions.
An inductive proximity sensor generates a high-frequency electromagnetic field at its face. When a metal object enters that field, it induces small currents in the metal that draw energy from the field, and the sensor detects that change and switches its output. Because it only responds to metal, an inductive sensor ignores dust, water, and non-metallic material around it, which makes it the workhorse for detecting metal targets, gears, cams, and machine parts.
A capacitive proximity sensor works differently: it senses a change in capacitance as an object approaches its face. Because capacitance changes with almost any material - metal, plastic, liquid, powder, wood - a capacitive sensor can detect non-metallic targets and is often used to sense the level of material through a container wall or to detect the presence of product. The trade-off is that this broad sensitivity also makes capacitive sensors more affected by moisture and buildup, so they need more care in placement.
Both types share the core advantages of non-contact sensing. With nothing physically touching the target, there are no mechanical parts to wear out, the switching is fast and repeatable, and the sensor can be fully sealed against the environment. Each type has a defined sensing range, and each requires the target to come within that range to trip, so the sensor is positioned with a deliberate gap to the object it watches.
Most proximity sensors have a solid-state transistor output rather than a mechanical contact, and that output comes in two common wiring conventions: PNP and NPN. A PNP sensor sources current - when it detects the target, it switches the positive supply through to the output, so the output goes high. An NPN sensor sinks current - when it detects the target, it connects the output to the negative side, pulling the output low. Both signal the same event; they differ in which side of the circuit the output connects to.
The distinction matters because the controller input must match the sensor. A PLC input card configured for sourcing sensors expects a PNP sensor, and one configured for sinking expects NPN. Wiring the wrong type to an input either does nothing or misbehaves, so specifying PNP versus NPN correctly is a routine but essential part of connecting a proximity sensor. In much of the world PNP sourcing sensors are the more common default.
Beyond output polarity, proximity sensors are also described as normally open or normally closed, which sets whether the output is active when the target is present or when it is absent. Choosing normally open versus normally closed is a matter of how the control logic and any fail-safe behavior are designed, so that a broken wire or lost sensor is interpreted safely.
In oilfield and facility applications, proximity sensors provide clean, reliable discrete inputs where physical contact would be a problem. They confirm position on moving equipment, detect the presence of parts or product, and are commonly used for speed and rotation sensing: a metal target passing an inductive sensor once per revolution generates a pulse train the controller can count to derive shaft speed or stroke rate. That makes them useful on pumping units and rotating machinery.
Fed into an RTU or PLC, each proximity sensor becomes a discrete or pulse input that the control system watches, and its ruggedness suits the outdoor, high-vibration, high-cycle conditions of the field where a contact switch would wear out. Sensors are chosen with the right sensing type, range, and output to survive the environment and to interface cleanly with the field controller.
In a cloud SCADA platform such as Merobix, that sensor data becomes remotely visible - a position confirmed, a count of strokes, a machine running or stopped - so an operator sees the state of remote equipment without a site visit. The non-contact nature of the sensor is part of why the signal stays trustworthy over long service life, which matters when the whole point is to reduce trips to the field.
An inductive sensor detects only metal objects by disrupting an electromagnetic field, so it ignores dust and moisture. A capacitive sensor detects almost any material - metal, plastic, liquid, or powder - by sensing a change in capacitance, which makes it useful for non-metallic targets and level sensing but more sensitive to buildup and moisture.
They describe how the sensor's transistor output connects when it detects a target. A PNP sensor sources current and switches the output high, while an NPN sensor sinks current and pulls the output low. The type must match the controller input - a sourcing input expects PNP, a sinking input expects NPN - or the signal will not read correctly.
Yes. If a metal target such as a bolt, cam, or gear tooth passes an inductive sensor once per cycle, the sensor produces a pulse each time, and a controller can count those pulses over time to calculate rotational speed or stroke rate. This is a common way to monitor rotating machinery and pumping units without contact.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.