Zhejiang Xinsenzheng Automation Co., Ltd.

Industrial Sensor Manufacturer OEM & Application Support Factory-direct Supply
Inductive proximity sensors installed on an automated inspection rig

Industrial sensor troubleshooting

How to Fix a Proximity Sensor Not Detecting Metal

Start with a known metal target and separate four observations: sensor power, sensor indicator, electrical output, and PLC input. Most failures come from the wrong target or distance, incorrect mounting, a PNP/NPN or NO/NC mismatch, damaged wiring, or a sensor that is unsuitable for the metal and environment.

  • Known-target sensing test
  • PNP, NPN, NO, and NC checks
  • Flush mounting, interference, and replacement

Inspection setup photo: Joongbae Kim, Wikimedia Commons, CC BY-SA 3.0.

Check 01 Use the exact model data

Read the supply, output, rated target, setting distance, and mounting instructions from the label and datasheet.

Check 02 Prove sensing first

Present a sufficiently large mild-steel target inside the model's stable operating range and watch the sensor indicator.

Check 03 Prove output second

If the indicator changes, test the output with the correct load and confirm PNP/NPN and NO/NC compatibility.

Check 04 Replace only with evidence

Confirm power, target, gap, mounting, wiring, and load before declaring the sensing element or output stage defective.

Control hazardous energy before touching the sensor or machine

Apply the site's energy-control procedure before adjusting brackets, cleaning a sensing face near moving parts, opening wiring, or entering a guarded area. A PLC stop command, sensor output, selector switch, or indicator is not an energy-isolating device.

Live voltage or output measurements should be performed only by qualified, authorized personnel under a controlled test procedure. If temporary energization is essential, clear tools and people, energize only for the test, then de-energize and reapply the required energy controls before continuing work. Follow local requirements and the machine-specific risk assessment.

Choose the observed symptom

Where does the detection chain stop?

Select what you see at the machine. The next checks change depending on whether the fault occurs at the sensor, its output, or the PLC.

Likely layer: supply, cable, or connector

Verify power at the sensor, not only at the cabinet

A dark indicator does not by itself prove a failed coil. Some models use the LED only for output state, while others have separate power and output indicators.

  1. Read the exact operating-voltage range and pinout.
  2. With an authorized live-test procedure, measure between the sensor's positive and 0 V terminals at the sensor connection.
  3. Inspect the fuse, terminal, cable, connector pins, splitters, and common return.
  4. Correct reverse polarity, excessive voltage drop, or an overloaded supply before retesting.

Next evidence: correct voltage at the sensor terminals under load.

Fast troubleshooting order

Use this five-step sequence before changing the sensor

Each step produces evidence for the next one and avoids replacing a healthy sensor while leaving the real wiring or application fault in place.

Identify

Record the full model, sensing principle, supply, output, switching function, connection, rated target, and installation type.

Inspect

With energy controlled, check the target path, gap, bracket, face, cable, connector, metal chips, impact, and surrounding metal.

Prove sensing

Use a known steel target at the specified stable set distance and observe the sensor indicator away from uncertain process variables.

Prove output

Under an authorized electrical test, verify supply and loaded output against the model's PNP, NPN, NO, NC, or 2-wire diagram.

Prove the system

Trace the state through the cable, PLC terminal, channel indicator, tag, program logic, machine cycle, and environmental extremes.

Target and distance

Why a sensor detects steel but misses aluminum, brass, copper, or stainless steel

An inductive sensor creates an alternating electromagnetic field at its active face. A metal target changes the oscillator through eddy-current losses and, for magnetic materials, permeability effects. The amount of damping depends on the sensor design as well as target material, alloy, size, thickness, approach direction, and distance.

The rated sensing distance is normally established with a defined standard target, often steel. It is not a guaranteed distance for every real machine part. Manufacturers publish a material correction or reduction factor for conventional inductive models. Factor-1 or all-metal models are designed to reduce this material dependence, but their exact assured distance still comes from the individual datasheet.

Do not use a generic aluminum factor as a final setpoint.

Typical values help explain a fault, but housing, shielding, coil design, alloy, and sensor family change the result. Use the factor and setting distance for the exact part number.

Target size matters too

A small bolt head, thin sheet edge, wire, tooth, or off-axis target can switch closer than the catalog's standard target. A larger target usually offers little extra range after it fully covers the useful field, while a smaller target can reduce range sharply. Recreate the real approach direction and speed during validation.

XSZ inductive proximity sensor with metal threaded housing and cable
Sensor diameter and housing appearance do not reveal the material correction factor. Check the complete model number and datasheet.
Reference target Steel Fe360 1.0

Typical reference value for a conventional inductive sensor.

Typical planning range Stainless steel 0.6-1.0

Composition and magnetic behavior can change the result significantly.

Typical planning range Aluminum 0.30-0.45

Verify the exact model or use a suitable factor-1 design.

Typical planning range Brass 0.35-0.50

Alloy and target geometry can move the switching point.

Typical planning range Copper 0.25-0.45

Use a measured validation margin rather than the upper estimate.

Typical ranges are engineering examples reported by industrial sensor manufacturers, not XSZ model specifications or universal limits.

Datasheet value Rated sensing distance
Exact model and metal Material correction factor
Starting estimate only Corrected switch-distance estimate
Then apply the manufacturer's stable setting distance.

Some product guides define the set distance at roughly 70% to 80% of the normal rated distance for the specified standard target. Treat the exact product value, assured operating range, tolerances, and application test as controlling. Do not automatically multiply every sensor by 0.8.

Installation faults

Check flush mounting, surrounding metal, the sensing face, and target alignment

A correct target can still disappear when the installation changes the electromagnetic field or the physical gap moves during the cycle.

Flush or shielded

Designed for embedding in metal

Internal shielding concentrates the field toward the front, allowing the active face to be mounted flush under the specified conditions.

  • Do not assume every long-range model is fully flush mountable.
  • Check front clearance, mounting nut position, and nearby moving metal.
  • Verify the target approaches the reference axis shown in the datasheet.
Non-flush or unshielded

Requires a metal-free zone

The field spreads around the active end, so surrounding steel can shorten, extend, latch, or destabilize the switching behavior.

  • Let the sensing face project as the model drawing requires.
  • Use the specified side, front, and mutual-interference clearances.
  • Do not copy a generic multiple of sensor diameter to every model.

Clean metal accumulation, but diagnose what it was doing

Nonmetallic oil or coolant is not automatically invisible to the complete installation, but a thin clean film normally does not behave like an extra millimeter of metal target. The larger problem is often metal powder or chips held in the fluid, a damaged seal that allows ingress, or a deposit that keeps the sensor continuously damped.

Record whether the sensor was stuck ON, failed to reset, or had a shortened practical gap before cleaning. Inspect the active face for cuts, impact marks, melted plastic, weld spatter, swelling, and abrasion. Never use the sensor as a mechanical stop.

Verify alignment through the whole cycle

Check bracket torque, shaft play, conveyor movement, actuator end position, target tilt, and vibration. Measure the minimum target overlap and maximum gap, not only the machine at rest. The switch and release points differ because inductive sensors have hysteresis; a target that barely turns the output ON may chatter as it moves away.

Industrial machine with metal components, brackets, and routed sensor cables
Inspect the complete moving geometry, not just the sensor at a stopped position. Photo: Mikhail Nilov via Pexels.

Output compatibility

Confirm PNP or NPN before testing the black output wire

Brown, blue, and black are common IEC-style conductor colors for 3-wire DC sensors, but the exact wiring diagram and connector pinout for the model always take priority.

Sourcing output

Typical 3-wire PNP sensor

Brown

Usually connects to the permitted positive DC supply.

Blue

Usually connects to 0 VDC or L-.

Black

Sources positive voltage to a load or sinking PLC input when active.

The load is typically between output and 0 V. Confirm input common, maximum load current, residual voltage, and protection behavior.

Sinking output

Typical 3-wire NPN sensor

Brown

Usually connects to the permitted positive DC supply.

Blue

Usually connects to 0 VDC or L-.

Black

Sinks current toward 0 V for a load or sourcing PLC input when active.

The load is typically between positive supply and output. Confirm that the PLC input type and common are compatible.

NO and NC describe switching function, not transistor type.

A PNP or NPN sensor can be normally open, normally closed, programmable, or complementary. Compare the timing chart with the PLC program. Use the NPN vs PNP Sensors guide and NO vs NC Sensor Output guide when the indicator changes but the control logic appears inverted.

Qualified technician using a multimeter at an industrial electrical enclosure
Measure at the sensor connection and PLC terminal under a controlled test procedure. Photo: Onics Energy via Pexels.

Electrical evidence

Test supply and output without creating a new fault

Do not assume every 3-wire sensor uses 10-30 VDC. Read the model label first. During an authorized live test, place the meter's red lead on the specified positive supply terminal and black lead on 0 V to check sensor voltage. Measuring at the power supply alone can miss cable drop, a loose connector, or a poor common.

Test the output with the intended load

An unloaded digital multimeter may display a voltage caused by leakage, diagnostic current, pull-up circuitry, or the PLC input. Test the sensor in the circuit defined by its output diagram and remain below the output current rating. Do not connect an arbitrary test lamp: its inrush current can overload a small transistor output.

Do not forget 2-wire behavior

A 2-wire electronic sensor is powered through its load. It has an OFF-state leakage current and an ON-state residual voltage. A high-impedance PLC input may remain falsely ON from leakage, while a low-voltage load may fail to energize because too much voltage remains across the sensor. Use the model's load-current window and the PLC input specification.

LED logic is model-specific.

An LED that is dark with correct supply does not prove a failed sensing coil, and an LED that is lit does not prove the output transistor or external circuit is healthy. Read the indicator definition and test each layer.

Trace the state

Use the LED, output voltage, PLC channel, and program tag as separate checkpoints

This matrix avoids the common mistake of calling every missing PLC signal a sensor detection failure.

Observation What it suggests Next evidence to collect
Observation No defined power or status indication
Suggestion Supply, polarity, cable, connector, protection, or model-specific LED behavior.
Next evidence Model pinout and actual voltage at the sensor under load.
Observation Indicator never changes with known steel
Suggestion Target outside set distance, target too small, wrong sensor type, surrounding metal, or sensor damage.
Next evidence Controlled target test using correct mounting and datasheet conditions.
Observation Indicator changes, output does not
Suggestion Wrong test reference or load, output short circuit, protection state, damaged output stage, or different indicator meaning.
Next evidence Loaded output measurement using the exact PNP/NPN timing and circuit diagram.
Observation Output changes, PLC channel does not
Suggestion Broken conductor, wrong M12 pin, input common mismatch, failed channel, or terminal problem.
Next evidence Compare the same state at sensor output, cable end, and PLC terminal.
Observation PLC channel changes, machine does not
Suggestion Wrong address, program inversion, interlock, sequence condition, timer, or machine-state logic.
Next evidence Trace the physical channel, PLC tag, cross-reference, and rung conditions.
A healthy sensor can look permanently OFF when the PLC input is incompatible.

Confirm sourcing versus sinking, input common, active-high versus active-low interpretation, NO/NC function, and the actual tag address before replacing hardware.

Intermittent and repeat faults

Look beyond the sensor when detection changes with the machine cycle

Temperature, electrical noise, nearby sensors, vibration, cable motion, and welding fields can reduce operating margin or create a false state.

Mutual interference

Two inductive sensors can interact when their oscillator fields overlap. Required spacing depends on sensor diameter, rated distance, flush/non-flush design, orientation, and frequency. Use the model's mounting diagram. Where close installation is unavoidable, select sensors with an interference-prevention function or offset frequencies.

EMI, power disturbance, and welding

Route sensor cables separately from motor, contactor, heater, VFD output, and welding conductors as the product and machine instructions require. Inspect cabinet bonding, cable shield continuity where applicable, surge suppression, and 24 VDC dips during solenoid or contactor operation. Near resistance welding, use a model specifically rated for weld-field immunity and spatter exposure.

Temperature and mechanical stress

Operation outside the rated temperature range can shift switching and release points and eventually damage electronics, resin, seals, or cable. Cable flexing near the molded exit, excessive connector strain, a loose bracket, or a target that moves with thermal expansion can create cycle-dependent dropouts.

Industrial control panel with terminal blocks and separated wiring
Trace the signal from field wiring to the PLC while noise-producing loads operate. Photo: Magda Ehlers via Pexels.
Only during motion

Inspect cable and alignment

Watch target overlap, bracket movement, connector strain, cable flex, and intermittent conductor continuity.

Only when loads switch

Inspect supply and EMC

Capture sensor voltage, output, grounding, routing, and transient suppression during the exact event.

Only when warm

Inspect temperature margin

Compare ambient and sensor temperature with the rating and recheck the target inside the stable set distance.

Make an evidence-based decision

When should you keep, replace, or upgrade the proximity sensor?

Choose the action that removes the verified root cause. A like-for-like replacement will not fix an unsuitable target, incompatible PLC input, or incorrect mounting pocket.

Keep and correct

The sensor passes a controlled test

Keep it when it switches reliably with the known target and correct load. Correct the target gap, bracket, mounting clearance, connector, PLC common, program logic, routing, or contamination that caused the field failure.

Replace

The sensor or lead is physically or electrically damaged

Replace it when the active face is cracked, abraded, swollen, melted, or impact-damaged; the integral lead is broken; ingress is visible; or it fails the manufacturer's known-target and loaded-output test with correct power and installation.

Upgrade the specification

The application exceeds the current sensor class

Consider factor-1 for varying metals, longer-range models for mechanical margin, shielded or non-flush geometry that matches the pocket, weld-field-immune versions, higher temperature or ingress ratings, robust cable, or IO-Link diagnostics.

Do not substitute a capacitive sensor only because the target is aluminum or copper.

Capacitive sensors can detect metal and nonmetal materials but introduce sensitivity to surrounding material, moisture, deposits, and adjustment. Compare a conventional inductive, factor-1 inductive, photoelectric, or other technology against the complete task. Start with the How to Choose Industrial Sensors guide.

Prevent repeat failures

Record the operating margin, not just the replacement part number

A short installation record makes the next fault faster to diagnose and prevents a maintenance replacement from changing a critical sensor characteristic.

01

Record the full model

Include sensing range, flush type, output, NO/NC function, connector, cable, temperature, and environmental rating.

02

Record the real target

Document alloy, dimensions, thickness, approach direction, speed, minimum overlap, and maximum operating gap.

03

Mark the stable set position

Measure ON and OFF points, then set the working position within the manufacturer's stable region with vibration allowance.

04

Inspect mounting and deposits

Include bracket movement, surrounding-metal clearance, target damage, active-face impact, and accumulated metal powder.

05

Trace the electrical chain

Label the output type, M12 pins, input common, PLC channel, tag, logic state, fuse, splitter, and junction locations.

06

Validate at operating extremes

Test minimum and maximum target position, cold and warm conditions, full machine noise, production speed, and cable movement.

Need help choosing a sensor that reliably detects your metal target?

Send the current sensor model, target alloy and size, minimum and maximum gap, mounting drawing, supply, PNP/NPN and NO/NC requirement, PLC input, temperature, contamination, and switching speed. XSZ can help narrow the correct sensor class before replacement.

Review My Detection Problem

Frequently asked questions

Proximity sensor not detecting metal FAQ

Short answers for common field symptoms and replacement decisions.

Why does my inductive proximity sensor detect steel but not aluminum?

Conventional inductive sensors are often rated with a steel target. Aluminum usually has a lower model-specific correction factor, so the usable switch distance can be much shorter. Move the target into the exact sensor's stable set distance or select a suitable factor-1 sensor after validating the alloy, size, and mounting.

How can I test an inductive proximity sensor without a PLC?

Use the manufacturer's circuit, a current-limited supply within the exact operating range, and a compatible load for the PNP, NPN, or 2-wire output. Present a known steel target inside the set distance and measure the loaded output. Do not use an arbitrary test lamp or guess the connector pins.

What voltage should I measure at a 3-wire proximity sensor?

Measure the voltage specified on the exact model, at the sensor terminals and under load. Many industrial 3-wire sensors use 24 VDC systems, but the permitted range is product-specific. With common color coding, red meter lead goes to positive or brown and black meter lead to 0 V or blue, after the pinout is confirmed.

Why does the sensor LED switch but the PLC input stay OFF?

The sensor may be detecting while the output circuit is incompatible or disconnected. Check PNP versus NPN, NO versus NC, PLC input common, black output conductor or M12 pin, loaded output voltage, input terminal, physical channel indicator, tag address, and program logic.

What is the difference between flush and non-flush proximity sensor mounting?

A flush or shielded inductive sensor is designed for embedding in metal under its specified conditions. A non-flush or unshielded sensor normally provides more range but needs a metal-free zone around the active end. Use the model drawing because clearance and mutual-interference distances are not universal.

Can metal chips on the sensing face stop detection?

Yes. Accumulated metal powder or chips can keep the field damped, prevent reset, or reduce useful margin. Control energy, clean the face using the approved method, inspect for damage or ingress, then repeat the known-target test and correct the source of the contamination.

How do I know the proximity sensor itself is bad?

Suspect the sensor after the exact supply, pinout, load, known target, set distance, mounting clearance, cable, and connector are verified and the sensor still fails its manufacturer-defined test. Physical face damage, ingress, a broken integral lead, or a loaded output that cannot switch are stronger replacement evidence than LED state alone.

Should I replace an inductive sensor with a capacitive sensor for non-ferrous metal?

Not automatically. A capacitive sensor can detect metal but may also respond to nearby nonmetals, moisture, or deposits. Compare a conventional inductive sensor at a shorter gap, a factor-1 inductive sensor, a photoelectric sensor, and other technologies against the real target, environment, speed, and mounting space.

Select the fields to be shown. Others will be hidden. Drag and drop to rearrange the order.
  • Image
  • SKU
  • Rating
  • Price
  • Stock
  • Availability
  • Add to cart
  • Description
  • Content
  • Weight
  • Dimensions
  • Additional information
Click outside to hide the comparison bar
Compare