Shielded vs Unshielded Proximity Sensors: When to Use Each

What is the difference between shielded and unshielded proximity sensors?

For an inductive proximity sensor, shielded usually means flush-mountable, while unshielded usually means non-flush-mountable. The distinction concerns the sensing head and how nearby metal affects detection. It is not a choice between a shielded and an unshielded cable.

Both types detect metal without contact. In a conventional shielded design, shielding around the coil reduces the field extending to the sides. An unshielded design has a wider field around its sensing end, so a bracket or machine frame can influence it as well as the intended target.

Original conceptual side sections, not to scale. Dashed boundaries are explanatory, not installation dimensions or measured sensing fields. The hero is a generated product illustration, not an exact-model photograph.

A projecting plastic head is a useful visual clue, but not a dependable way to classify every sensor. Read the mounting designation in the product data. Cable screening, EMC performance and weld-field immunity are separate specifications.

Which mounting type should you choose first?

Start with the metal around the sensing face, then check the required detection gap. A sensor with an attractive range rating is not a good fit if the machine cannot provide its installation clearance.

When is a flush sensor the practical choice?

Consider a target flag passing a steel fixture: keeping the sensor face level with the permitted mounting surface may reduce how far the head protrudes into the moving part's path. This suits compact tooling when the available sensing distance is sufficient. It does not turn the sensing face into an impact stop.

When does a non-flush sensor earn its extra space?

Consider a metal carrier passing a stand-off bracket. If the head can project and remain clear of the frame and guards, a non-flush model may provide useful reach. Include the bracket, fasteners and adjacent equipment in the space check, not just the target-to-face gap.

A flush sensor can also be used in an open mounting arrangement when its instructions and range suit the application. Open space does not make a non-flush sensor mandatory.

Does an unshielded sensor always detect farther?

No. Non-flush versions often have a longer rated range within a comparable conventional family, but there is no universal multiplier between the two types. Newer designs and extended-range variants can change that comparison.

For example, Pepperl+Fuchs' installation guide lists typical 18 mm designs with 5 mm switching distance for flush installation and 8 mm for non-flush installation. Those are examples in that guide, not specifications for every M18 sensor or for xsz sensor products. Compare the full ordering codes before applying a range claim to a purchase.

Compare usable distance, not just the headline number

A published sensing distance is tied to a defined target and test method. Your small screw head, aluminum part, curved carrier or tilted flag may produce a different result. Look for the manufacturer's recommended set distance or assured operating distance, with its conditions, and the target-size or material data relevant to your part.

Check both ends of the machine's movement: the largest gap is the difficult detection condition; the smallest gap is the collision check. Then confirm where the target must release the sensor again. A successful approach at one carefully adjusted gap does not establish reliable operation throughout the tolerance range.

If the target travels sideways across the face, its path through the sensing area matters too. A head-on distance rating does not, by itself, identify the switching position along that path.

How much metal clearance does the installation need?

Use the installation drawing for the selected model and mounting arrangement. There is no single clearance that covers the surrounding metal, another sensor and the intended target.

Flush is not the same as recessed

In a normal flush installation, the sensing face sits at the permitted metal surface. Sinking it below a steel lip or placing metal over the face creates a different geometry. That nearby metal can influence detection; it is not a way to add protection without changing performance.

Use a recessed arrangement only if the manufacturer explicitly supports that geometry. Likewise, do not push a quasi-flush sensor fully into the bracket because its name contains “flush.” ifm, for example, describes quasi-flush products as an almost-flush alternative with their own installation depth.

Read each dimension from its own reference points

• Head projection: how far the specified part of the sensor must extend beyond the mounting surface.
• Surrounding-metal clearance: the free space required beside or in front of the active end. Check whether the drawing specifies a hole diameter, radius or edge gap.
• Adjacent-sensor spacing: the separation for the shown side-by-side or opposing orientation. Confirm whether it is measured between faces, edges or centerlines.

A rule such as “3 × diameter” is incomplete without its drawing and applicable sensor family. Do not transfer it to a different dimension or assume the same letter means the same measurement in another manual.

For a tight sensor array, ask about documented reduced-interference or alternate-frequency variants rather than reducing the prescribed spacing by trial and error. Turning a neighbor off does not remove its metal housing from the installation.

Can you replace a shielded sensor with an unshielded model?

Sometimes, but it is not a drop-in replacement merely because the thread, voltage and output type match. The new sensor must also suit the bracket geometry, target path and electrical interface.

Also compare body length, connector orientation, cable clearance, pin assignment, load requirements and switching performance. The right mounting class does not automatically make the rest of the replacement compatible.

Why does the sensor work in free air but fail in its bracket?

The mounting metal may be influencing the sensing head, but that observation alone does not prove the cause. An adjacent sensor, a changed target path or an electrical connection can produce a similar symptom.

Before changing a mount or wiring, isolate the machine and prevent unintended motion. A qualified person can compare conditions on a controlled test fixture; do not remove a sensor from a running machine to perform these checks.

1. Confirm what “ON” means. Check the NO/NC function and the manual's LED description. An energized output is not always the target-present state.
2. Separate metal influence from active interference. With the target geometry held constant, compare the suspect bracket arrangement with the documented mounting arrangement. For a neighbor test, keep its housing in place while changing its powered state on the controlled fixture. A change is a diagnostic clue; common-supply effects still need to be excluded.
3. Compare the sensor signal with the controller input. If the sensor output is stable but the PLC indication is not, investigate the interface, connection and input processing instead of repeatedly moving the sensor.

Record the target-present and target-absent results for each condition. If behavior changes when you correct the mounting clearance, retain that geometry and verify the complete operating range. A one-off LED change is not a validated production setting.

What should you confirm before ordering and installing it?

A useful selection record connects the proposed part number to the actual machine layout. Keep it short enough that purchasing, installation and maintenance can use the same information.

• Part and mounting: full ordering code, drawing revision, permitted mounting depth and all relevant metal and sensor clearances.
• Target and movement: material, dimensions, approach direction, minimum and maximum face gaps, and required detection and release positions.
• Interface and environment: supply, PNP/NPN and NO/NC function, pinout, load, required switching behavior, temperature and exposure conditions.
• Installed check: detection and release with the finished bracket, guards and all sensors operating, including the expected target and position variations.

Agree the acceptable switching positions and any missed or unwanted detection limits before the trial. Record the tested arrangement so a later replacement does not silently change it. Ordinary process-sensor selection is not validation of a personnel-protection safety function.