Inductive Sensor

What is an Inductive Sensor

An inductive proximity sensor is a non-contact device that signals when an electrically conductive metal object comes within a defined distance of its active face. It belongs to the broader proximity-switch family alongside capacitive, magnetic, photoelectric, and ultrasonic types, but it is unique in responding only to metals and in doing so through a purely electromagnetic mechanism with no light, no sound, and no mechanical plunger. In a typical machine, dozens of these sensors confirm that a cylinder has reached end of stroke, that a part is present in a fixture, that a turret has indexed, or that a shaft is turning, feeding binary status straight to a PLC.

The operating principle is the eddy-current killed oscillator, often abbreviated ECKO. The sensor contains four functional blocks: a ferrite-cored coil that forms the inductive element of an LC resonant circuit, an oscillator that sustains an alternating field at that coil, a demodulator that measures the oscillation amplitude, and an output stage that drives the switching transistor. With no metal nearby, the oscillator runs freely. When a conductive target enters the field, the field induces circulating eddy currents in the target; those currents dissipate energy and, by Lenz's law, set up an opposing field that damps the oscillator. As the target moves closer, damping deepens until the amplitude crosses a threshold and the output switches. Withdraw the target and the oscillator recovers, switching the output back at a slightly farther point because of designed-in hysteresis.

Because detection is field-based, an inductive sensor never touches the target. That gives it the defining advantages that made it ubiquitous: no contact wear, so service life is governed by electronics rather than mechanical fatigue; immunity to dirt, oil, dust, and non-metallic debris in the gap, since only conductive metal is sensed; tolerance of vibration and shock that would destroy a mechanical limit switch; and fast, bounce-free switching suited to high-speed counting. The limitations are equally definite: it senses metals only, not plastic, glass, liquid, or wood; its range is short, from under a millimeter to a few tens of millimeters; and its rated distance depends on the target alloy, which is the single most common source of commissioning surprises.

The technology has a long industrial pedigree. The eddy-current effect was described by Leon Foucault in 1855, and the LC-oscillator proximity switch was commercialized through the 1960s as solid-state electronics replaced electromechanical limit switches on production lines. Standardization followed: IEC 60947-5-2, the international standard for proximity switches under low-voltage switchgear, fixed the definitions of nominal sensing distance, the standard target, and the operating-distance tolerances that every reputable datasheet now quotes. Today the global installed base runs to billions of units, and the barrel form factors, M8, M12, M18, and M30 threaded bodies, are produced as interchangeable commodities by every major automation supplier while specialty variants serve welding, high temperature, and hygienic duties.

Four engineering attributes determine whether a given inductive sensor fits an application: the sensing distance and how it derates with target material, the housing and ingress protection against the working environment, the electrical interface (voltage, output type, and wiring) matched to the controller, and the switching frequency relative to how fast targets pass. The chapters that follow take each in turn, but the recurring theme is that the round catalog number, the nominal distance, is a starting point and not a design value; the assured distance and the material factor are what you actually engineer against.

Shielded and Unshielded Types

The first and most consequential classification is whether the sensor is shielded (also called flush or embeddable) or unshielded (non-flush or non-embeddable). The distinction is mechanical: a shielded sensor wraps a metal sleeve around the coil so the magnetic field projects only forward from the active face, while an unshielded sensor leaves the coil sides open so the field bulges outward as well. That single construction choice drives both the achievable range and the mounting rules, and getting it wrong produces either lost range or chronic false triggering.

A shielded sensor can be threaded into a steel mounting block so that its face sits flush with the surrounding metal, because the surrounding metal lies outside the forward-only field and does not damp the oscillator. This is invaluable in dense fixtures and machine frames where the sensor must be protected and space is tight. The penalty is range: confining the field shortens the sensing distance. As a representative figure, a shielded M12 sensor reaches roughly 2 mm against the standard steel target, where the equivalent unshielded version reaches roughly 4 mm.

An unshielded sensor trades mounting freedom for distance. Because the field also projects from the sides, the head must protrude from any metal it is mounted in and must keep a clear non-metallic zone around it; manufacturers typically specify a free diameter of about three times the barrel size and a free axial zone in front of three times the sensing distance. Mount an unshielded sensor flush in metal and the frame itself damps the oscillator, producing a permanent or erratic ON state. Used correctly, the longer range eases mechanical tolerance and lets the sensor see past minor obstructions.

Across both families the standard form factor is the threaded cylindrical barrel, with M8, M12, M18, and M30 covering the bulk of industrial use, supplemented by miniature M4, M5, and M6 bodies and by rectangular block housings for long-range or high-density layouts. The approximate nominal sensing distances below show how range scales with body size and shielding; exact figures vary by series and must be read from the specific datasheet.

Long-range variants extend these figures further, with extended-range M12 and M18 bodies reaching roughly 6 mm and 12 mm respectively, at the cost of slightly tighter mounting and lower maximum switching frequency. When range is genuinely insufficient even from an unshielded long-range part, the alternative is to step up barrel size or switch to a different sensing family altogether, since pushing an inductive coil far beyond its geometric range degrades repeatability.

Sensing Technologies and Variants

While almost all inductive proximity sensors share the ECKO oscillator core, the family has diversified into purpose-built variants that solve specific field problems: sensitivity to target alloy, immunity to welding fields, analog distance output, and intrinsic safety. Understanding which variant maps to which problem prevents both over-specifying a costly special where a commodity part suffices and under-specifying a commodity where a special is mandatory.

The standard ferrous sensor is the commodity workhorse: rated against mild steel, tuned for the lowest cost, and adequate wherever the target is steel and the environment is ordinary. The remaining variants exist because real applications break one of those assumptions.

A Factor 1 sensor attacks the material-derating problem. Ordinary sensors lose range on aluminum, brass, and copper because those metals damp the oscillator less than steel does; a Factor 1 design uses a multi-coil or air-coil sensing system so that ferrous and non-ferrous metals are detected at the same nominal distance, with a correction factor of 1 across the board. This is essential where the target alloy varies on the line, where aluminum or copper must be sensed at full range, and as the starting point for welding-grade sensors. Pepperl+Fuchs, Balluff, Contrinex, and Baumer all publish Factor 1 lines.

A weld-field immune sensor targets the resistance-welding cell, where the kiloampere AC weld currents radiate intense magnetic fields that saturate or false-trigger an ordinary inductive coil. Weld-immune designs use an air-coil system and shielding electronics to reject those fields, and they are commonly combined with Factor 1 sensing (since automotive body shops weld steel and aluminum) and a PTFE-coated active face that resists the adhesion of molten weld spatter and slag. Sacrificial coatings and thread-protected bodies extend life in the spatter zone.

An analog-output sensor abandons binary switching and instead delivers a continuous signal, typically 0 to 10 V or 4 to 20 mA, proportional to the target distance across a measuring window. This turns the inductive principle into a short-range non-contact displacement gauge for tasks such as strip-edge guiding, eccentricity and runout checking, valve-position feedback, and fill-level sensing of metal contents. Resolution is high within the window, but the usable span is short and the analog output, like the switching type, still derates with non-ferrous targets unless a Factor 1 analog part is chosen.

A NAMUR sensor is a two-wire device built to EN 60947-5-6 with a defined low-energy current characteristic, intended for connection to an isolating switch amplifier in the safe area. Because it carries minimal energy, it is the standard inductive building block for intrinsically safe loops in hazardous (ATEX or IECEx) areas, where the amplifier converts the small current change into a clean binary signal and also detects wire-break and short-circuit faults. Selecting a NAMUR sensor without its matched amplifier is a common specification error.

Beyond these, ring and slot sensors detect small metal parts passing through an aperture (useful for fastener counting and small-part feeding), and full-metal stainless variants integrate the coil behind a closed metal face for extreme washdown and pressure-cleaning duty. The selection logic in every case is the same: start from the cheapest standard part, then upgrade only for the specific stressor, material, weld field, hazardous area, or analog requirement, that the standard part cannot meet.

Target Materials and Correction Factors

The most important number on an inductive sensor datasheet, the nominal sensing distance Sn, is defined against one specific target and one specific target only. Under IEC 60947-5-2, the standard target is a square plate of mild steel (grade Fe 360, also written St37), 1 mm thick, with a side length equal to the larger of the active-face diameter or three times Sn. Every catalog distance figure assumes that plate. The moment your real target is a different alloy, a different thickness, or a smaller size, the achievable distance changes, and the change is large enough to break a marginal design.

The dominant effect is alloy. Steel damps the oscillator strongly because it is both conductive and ferromagnetic; non-ferrous and stainless metals damp it less, so the real distance is Sn multiplied by a reduction (correction) factor below 1. The table below lists representative factors; exact values vary by sensor series and alloy temper, so the manufacturer correction chart is the authority for any critical gap.

The practical reading of this table is that a sensor sized for steel will switch at less than half its rated distance on aluminum or copper. A 12 mm rated sensor sensing aluminum switches at roughly 4.8 mm, so a mechanical gap set at 8 mm, comfortable for steel, never trips on aluminum. The two standard remedies are to either size the sensor and the gap explicitly against the derated distance, or to specify a Factor 1 sensor whose correction factor is 1 for all metals and which therefore removes the guesswork entirely.

Three secondary effects refine the target picture. Target size: a target smaller than the standard plate reduces the distance, so small parts, thin wires, and the edges of a target all read short; conversely a target larger than the standard plate gives no extra range. Target thickness: for ferrous targets, 1 mm is enough to reach full rating, but for non-ferrous foils the eddy currents need adequate thickness, and very thin aluminum or copper (skin-depth limited) reads shorter still. Plating and coatings: a thin conductive plating over a different base, such as galvanized or chrome plating, shifts the effective factor toward the surface metal, which can help or hurt depending on the layer.

Two installation factors complete the chapter. Surrounding metal damps the field just as a target does: a shielded sensor tolerates flush mounting in metal, but an unshielded sensor needs the free zones described in Chapter 2, and a target approaching from the side rather than the face will switch unpredictably. Mutual interference arises when two sensors are mounted too close: facing sensors and side-by-side sensors must respect the manufacturer minimum spacings, often expressed as multiples of the active diameter, or their oscillators beat against each other and produce erratic switching. Both effects are deterministic and are resolved by reading the mounting clearances in the datasheet rather than by trial and error in the field.

Key Specification Parameters

An inductive sensor datasheet typically lists 15 to 25 parameters, but a manageable set governs nearly every selection: the family of operating distances, hysteresis and repeatability, switching frequency, the electrical interface, environmental ratings, and built-in protection. Each is decoded below in the language of IEC 60947-5-2.

Operating distances are a family of four defined terms, not a single number, and confusing them is the classic novice error. Sn (nominal/rated) is the round catalog figure with no tolerances applied. Sr (real/effective) is the measured distance of an individual unit at 23 degrees Celsius and rated voltage, permitted to lie between 0.9 and 1.1 times Sn. Su (usable) folds in the full temperature and supply-voltage spread and is permitted to lie between 0.9 and 1.1 times Sr. Sa (assured operating distance) is the conservative guaranteed figure, defined as 0 to 0.81 times Sn, within which the sensor is certain to detect the standard target under every permitted condition. Design the mechanical gap against Sa, never against Sn, and apply the material correction factor on top.

Hysteresis (differential travel, H) is the gap between the pick-up point as the target approaches and the slightly farther drop-out point as it withdraws, typically specified as 3 to 15 percent of Sr. It is a deliberate feature: without it, a target dwelling near the switch point under vibration would chatter the output on and off. Repeat accuracy (R) is the scatter of the effective distance between successive measurements under fixed conditions and is the metric that matters for precise positioning; quality sensors hold repeatability to a fraction of a percent of Sr.

Switching frequency is the maximum rate of target on/off cycles the sensor can resolve, from roughly 25 Hz on a long-range M30 up to several kilohertz on a small shielded M8. Verify it exceeds the actual target pass rate, for example gear teeth past the face or parts past on a fast index, with margin. Response and release time set the latency between target arrival and output change and matter in high-speed interlocks.

Electrical interface covers supply, output type, and wiring. Supply is most commonly 10 to 30 V DC. Output configuration is the pairing of switching element and action:

• PNP (sourcing) vs NPN (sinking): PNP switches the positive line to the load and is standard in Europe and most modern PLCs; NPN switches the negative line and persists in some Asian controls. The choice must match the PLC input card polarity.
• NO vs NC: normally open closes on target detection, normally closed opens on detection; NC is preferred in fail-safe interlocks where a broken wire should signal a fault.
• 2-wire, 3-wire, 4-wire DC: 2-wire is series-connected with off-state leakage and on-state voltage drop; 3-wire (supply +, supply -, output) is the industrial default; 4-wire adds a complementary or second output.
• IO-Link: a point-to-point digital layer over the standard 3-wire connection that adds remote parameterization, diagnostics, and process data through an IO-Link master, the trend in smart-factory builds.
• Analog and NAMUR: 0 to 10 V or 4 to 20 mA distance output for measurement, and low-energy NAMUR two-wire output for intrinsically safe loops.

Environmental and ingress ratings determine survival. Housing is nickel-plated brass for general factory air or stainless steel (303 or 316L) and full-metal one-piece bodies for washdown and corrosive duty. Ingress protection follows IEC 60529: IP67 (dust-tight, temporary immersion) is the practical minimum, IP68 for continued immersion, and IP69K for high-pressure, high-temperature cleaning in food and beverage. Ambient temperature commonly spans -25 to +70 degrees Celsius, extended to +85 degrees Celsius for stainless variants and further for special high-temperature designs. Connection is by integral cable or by M8 and M12 quick-disconnect connectors.

Built-in protection and diagnostics round out the list: short-circuit protection on the output, reverse-polarity protection on the supply, transient and overvoltage suppression, and an LED status indicator (and often a separate power LED) for at-a-glance commissioning. For hazardous areas, the relevant ATEX or IECEx certification applies; for safety functions, the documented SIL or PL rating of safety-listed models is mandatory.

Selection Decision Factors

To convert the preceding five chapters into a specific part number, follow the ordered sequence below. Most selection failures come not from one wrong answer but from deciding a downstream parameter before an upstream one is settled, for example fixing the barrel size before confirming the target alloy. These steps double as a fixed RFQ template.

1. Target material and size: Identify the alloy first. If it is mild steel and constant, a standard sensor is fine; if it is aluminum, brass, copper, stainless, or mixed, either derate by the Chapter 4 factor or specify a Factor 1 sensor. Confirm the target is at least the size of the standard plate, or expect reduced range.
2. Required sensing distance and mounting: Set the mechanical gap, then work back to the needed distance against Sa (0.81 times Sn) with the material factor applied. Decide shielded (flush in metal, shorter range) or unshielded (longer range, free zones required) from how the sensor must be mounted.
3. Barrel size and form factor: Choose M8, M12, M18, or M30 (or miniature, ring, slot, or block) to deliver the derated distance with margin while fitting the available space. Larger barrels buy range; smaller barrels buy density.
4. Switching speed: Confirm the rated switching frequency exceeds the actual target pass rate (teeth, slots, or parts per second) with headroom, and check response and release times for fast interlocks.
5. Electrical interface: Match supply voltage, PNP or NPN to the PLC input polarity, NO or NC to the fail-safe logic, and 2-, 3-, or 4-wire (or IO-Link, analog, NAMUR) to the control architecture.
6. Environment and ingress: Pick housing material and IP rating for the worst-case exposure (coolant, washdown IP69K, weld spatter, vibration), and confirm the ambient and target temperatures fall inside the rated window.
7. Certifications and protection: Add ATEX or IECEx for hazardous areas (usually NAMUR plus amplifier), SIL or PL for safety functions, and verify short-circuit, reverse-polarity, and transient protection plus an LED indicator are present.
8. Connection and total cost: Choose integral cable or M8/M12 connector for fast field replacement, then weigh unit price against downtime cost. A connectorized, correctly derated sensor that never false-trips is cheaper over its life than a marginal part that stops the line.

One dimension that buyers routinely overlook is serviceability and standardization. Inductive proximity sensors are high-population, high-replacement-rate parts, so favoring a few standard barrel sizes and connector types across a plant, stocking spares, and choosing suppliers with reliable local availability cuts mean-time-to-repair far more than a marginal spec advantage on any single sensor. Established manufacturers, including Pepperl+Fuchs, Balluff, ifm electronic, Turck, Omron, SICK, Eaton, Schneider Electric, Carlo Gavazzi, Baumer, and Contrinex, publish full correction charts, mounting clearances, and CAD models, and maintain interchangeable form factors that make field substitution straightforward. For mature lines, that ecosystem support often outweighs a few millimeters of extra range on a datasheet.