A sensor fault can be the sensor, the target, the wiring, the input, the setup, or the timing. Separate them.
SURVIVAL RULE: Check target, power, signal, cable, and input separately.
Separate target problems from sensor problems
A healthy sensor can report the wrong state because the target is missing, misaligned, too far away, reflective in the wrong way, covered in dust, or arriving at the wrong angle. Mechanical movement can shift brackets slowly until a once-reliable sensor reaches the edge of its sensing range.
A photoelectric sensor that fails only with black packaging may be a contrast or setup issue, not a broken sensor. An inductive sensor may fail after a bracket bends two millimeters away from the metal target.
Field habit: Before replacing a sensor, inspect target, gap, alignment, contamination, mounting rigidity, and teach/setup settings.
Common trap: Do not turn sensitivity to maximum as a permanent solution without understanding the effect on false detection.
Use LEDs as clues, not proof
Sensor LEDs are fast indicators of power and switching, but they do not prove the PLC receives the signal. A sensor can switch locally while a cable conductor is broken, a connector pin is corroded, or an input channel is damaged.
If the sensor output LED changes but the PLC input LED does not, the fault area is between the sensor output and input point. If both LEDs change but logic does not respond, investigate software interpretation or the wrong address.
Field habit: Observe the field LED and PLC input at the same time when possible.
Common trap: Do not stop troubleshooting at ‘the light comes on.’
Quick check
What should be true at this point in the sequence?
What evidence can prove or eliminate this section of the system?
What changed recently or only under the failing condition?
Intermittent sensors often point to wiring
Repeated vibration, flexing, oil, coolant, cable carrier movement, and connector strain create faults that appear only in certain positions. A sensor may test perfectly while the machine is stopped because the cable is no longer bending.
Watch live input state while safely flexing an isolated cable section where permitted, or observe the signal across repeated machine cycles. Pay special attention to cable exits, drag chains, connectors, and points where the cable rubs metal.
Field habit: Record which machine position triggers the fault. Position correlation is valuable evidence.
Common trap: Do not repeatedly wiggle energized wiring in a way that creates short-circuit or motion hazards.
Replacement must preserve function
A replacement sensor needs more than the same sensing principle. Check voltage, PNP/NPN, NO/NC, range, response speed, connector, pinout, environmental rating, mounting, and any special teach mode. Some machines depend on a very specific switching point or timing.
A longer-range sensor may seem like an upgrade but detect background metal and cause false triggers. A slower sensor may miss fast product.
Field habit: After replacement, verify the input during real motion and across the normal product range.
Common trap: Do not throw away the failed sensor before recording its exact part number and failure evidence.
Quick check
What should be true at this point in the sequence?
What evidence can prove or eliminate this section of the system?
What changed recently or only under the failing condition?
Field Exercise
Select one sensor and verify target, power, output, cable, and PLC input while the machine is healthy.