Inductive Proximity Sensor Circuit Design

An inductive proximity sensor detects a metal target without touching it, using a coil that generates an alternating magnetic field and a circuit that notices when the field loses energy to a nearby conductor. It works in dust, oil and water where optical sensors fail, and it has no moving parts to wear out. The design is a small exercise in oscillator stability.

How Inductive Sensing Works

An oscillator drives a coil with an alternating current, producing an alternating magnetic field in front of the sensor. When a metal target enters the field, the field induces circulating currents in the target, and those currents dissipate energy. The oscillator sees the extra loss as a reduction in amplitude.

The detection circuit monitors the amplitude of the oscillation. As the target approaches, the amplitude falls, and at a defined level the output changes state. The relationship between the distance and the amplitude change is not linear, so the switching point is defined by the threshold rather than by the amplitude curve.

The same principle is used in a linear position sensor, where the amplitude is measured rather than compared with a threshold. In that case the calibration and the temperature stability matter more, because the measurement depends on the absolute amplitude rather than on a comparison with a fixed reference.

The Oscillator and Sensing Coil

The oscillator is usually a tuned circuit with the sensing coil as its inductor and a capacitor to set the frequency. Frequencies between a hundred kilohertz and a few megahertz are common, with the higher frequencies giving a better response to small or thin targets and the lower frequencies giving a longer sensing distance.

oscillator coil design determines the sensing distance. A larger coil produces a larger field and a longer range, but it also has more capacitance to its surroundings and is more sensitive to nearby metal other than the target. The coil is usually printed on the board or wound on a small ferrite core, and the choice affects both the range and the mechanical packaging.

The oscillator has to start reliably and continue oscillating with a stable amplitude. The amplitude is set by the balance between the energy supplied by the active device and the losses in the circuit, and a design with too little margin will fail to start at low temperature or with a damped coil. The margin is checked by measuring the amplitude over temperature.

Eddy Currents and the Detection Threshold

eddy current is the name for the circulating current induced in the target. Its magnitude depends on the conductivity and the permeability of the material, and on the frequency of the field. A target with high conductivity, such as copper or aluminium, produces strong eddy currents, while a ferromagnetic target also concentrates the field and changes the coil inductance.

Because two different mechanisms are at work, the response depends on the material. A ferrous target reduces the inductance and increases the loss, while a non ferrous target mainly increases the loss. A sensor calibrated for steel will have a shorter range on aluminium and a different one on brass, and the reduction factor is quoted on the datasheet for each material.

The detection threshold is set so that the output switches at a defined distance from the target. It is usually adjustable, either by a potentiometer or by a register, and the adjustment range has to cover the tolerance of the coil and the spread of the oscillator amplitude between units. A design with no adjustment requires very tight component tolerances.

Sensing Distance and Target Material

sensing distance is quoted for a standard target, usually a square of mild steel one millimetre thick with a side equal to the diameter of the sensing face. Anything smaller or thinner than the standard target reduces the distance, and the reduction is significant for targets thinner than about half a millimetre.

The mounting environment matters as much as the target. A sensor mounted flush in a metal panel loses part of its field to the panel unless it is designed for flush mounting, and one mounted close to another sensor may interact with it. The datasheet usually specifies a minimum distance between adjacent sensors and a minimum clearance from surrounding metal.

The repeatability of the switching point is the parameter that matters in a positioning application. It is typically a small fraction of the sensing distance, and it depends on the stability of the oscillator amplitude rather than on the absolute accuracy of the distance. A stable oscillator with a well defined threshold gives repeatable switching even when the absolute distance is not precise.

Hysteresis and Switching Frequency

hysteresis makes the release distance longer than the operate distance, so a target that is moving slowly or vibrating near the threshold does not produce a burst of output transitions. Without it, a sensor watching an object that stops exactly at the switching point will chatter, and any counter attached to the output will accumulate spurious events.

The amount of hysteresis is a compromise. A large window makes the sensor robust against vibration but reduces the repeatability of the switching point, because the operate and release points are further apart. In a positioning application the window is kept small, and mechanical damping or a faster target movement is used instead.

The maximum switching frequency is related to the oscillator and to the output stage. Each detection requires the oscillator amplitude to change by an amount the comparator can resolve, which takes a number of cycles, and the output stage has to slew. A sensor specified at several kilohertz is responding to a toothed wheel rather than to a single target.

Output Stage and Interface

The output is usually a transistor that can sink or source a defined current, with protection against reverse connection and against an inductive load switching off. A short circuit on the output is common in a machine installation, so the output stage should tolerate it or be protected by an external element.

Two wire sensors draw their operating current through the load, which means a small leakage current flows when the output is off. In a circuit with a sensitive input that leakage can be interpreted as an on state, and the load has to be chosen so that the off state voltage stays below the input threshold. Three wire sensors avoid the problem but need an additional conductor.

The interface to the controller should include a filter and, where the cable is long, a series resistor. A sensor cable running alongside a motor cable picks up interference, and the general methods for finding and removing that coupling are described in our article on EMI suppression design, while the specific question of choosing a suppression component is covered in ferrite bead selection.

Layout and Coil Design

If the coil is printed on the board, its geometry determines the field shape and the sensing distance. A spiral on the outer layer with a ferrite backing concentrates the field in front of the sensor, while a coil without backing radiates in both directions and responds to metal behind it as well as in front. The backing is part of the design rather than an accessory.

Keep the oscillator components close to the coil and away from the output stage, because the output switching current is large compared with the oscillator current and can shift the amplitude. The oscillator is a sensitive analogue circuit and belongs on the quiet side of the board partition.

The board layout also has to consider the metal around the sensor. A ground plane under the coil reduces the field and the sensing distance, and a metal fixing screw close to the coil has the same effect. Both should be evaluated on a prototype, because the effect is not easy to predict from the geometry alone.

Verification and Common Faults

Verify the sensor by measuring the output transition distance for a standard target, approaching and then receding, and recording both points. The difference is the hysteresis, and the midpoint is the nominal switching distance. Repeating the measurement over temperature shows how much the amplitude drift moves the switching point.

A sensor that works with a steel target and not with aluminium has a sensing distance problem rather than a fault, and the reduction factor should be applied in the specification. A sensor that detects a target behind it usually has no ferrite backing, or the backing is not correctly positioned.

An output that chatters when a machine is running, but not when it is stationary, is picking up vibration or electrical noise. Comparing the behaviour with the machine running but the sensor mounted on a separate bracket separates the mechanical and electrical causes. Consistent switching distance is the property to verify in production, and the inspection practices that catch a drifting unit are described in our guide to judging PCB quality.

FAQ

How far can an inductive sensor detect a target? Typically between a few millimetres and about twenty millimetres, set by the coil diameter and the oscillator power. Longer ranges are possible but require a larger sensing face.

Why does a smaller target reduce the sensing distance? The eddy currents in a small target dissipate less energy, so the change in oscillator amplitude is smaller. Targets thinner than half a millimetre reduce the distance noticeably.

Can I use an inductive sensor through a metal wall? Not through a ferromagnetic wall, which shields the field. A non magnetic wall such as stainless steel reduces the distance, and plastic has almost no effect.