PLC Sensor Input Explained: Sinking, Sourcing and Wiring

What does a PLC sensor input actually do?

A PLC sensor input converts an electrical signal into information the controller can use. A discrete input reports a state, such as part present or part absent. An analog input reports a measured value. The sensor output and the PLC input must agree on more than the supply voltage: signal type, current path, electrical limits and timing all matter.

Think of the connection as a chain: target → sensor output → input circuit → input data → program action. A detection LED confirms only part of that chain. It does not prove that the PLC has received the correct event.

A dry contact does not generate its own sensing voltage; the input circuit needs the supply arrangement specified by its manufacturer. AC inputs are another interface again. Neither should be selected from a generic “24 V sensor” description.

Does your sensor need a sinking or sourcing input?

For conventional three-wire DC sensors, a PNP output sources current into a sinking PLC input. An NPN output sinks current from a sourcing PLC input. The names describe opposite ends of the same current path, not two matching labels.

Which common terminal belongs to this channel?

Find the exact input channel and its associated COM, M or L+ terminal in the module diagram. Labels and grouping vary. A configurable module may allow either polarity for a group, without allowing arbitrary PNP/NPN mixing within that group. Do not move a shared common without checking every connected channel.

For a replacement, compare the full sensor suffix, output diagram and cable pinout with the input-module diagram. Brown, blue and black are common DC sensor wire colors, but a color convention is not a substitute for that check.

Does NPN mean the PLC reads OFF when the voltage is low?

No. An active NPN output can pull its signal terminal toward 0 V while current flows through a compatible sourcing input. That input can then report logic 1. “Low voltage at the wire,” “output ON” and “program condition true” are different statements.

NO/NC describes when the output activates relative to the detection condition; PNP/NPN describes its electrical behavior. Confirm both states in the actual program, including any inverted contact or tag logic.

Why can a 24 V sensor still fail to switch the input?

“24 V” is a nominal system description, not a complete compatibility rating. The input has specified ON and OFF regions. The sensor must drive the connected input into each required region across supply variation, cable losses and operating conditions.

Check the loaded ON state—not just the supply terminals

Compare the module's guaranteed ON requirements with the sensor's output voltage drop and current capability. Include the minimum available supply and wiring losses. A sensor's maximum load current is a limit on what it can drive; it is not the current it automatically sends into every PLC input.

Check the OFF region separately

An input may have an undefined region between guaranteed OFF and guaranteed ON. A reading in that region is not a reliable switching state. Check the specified current as well as voltage criteria where provided, and take measurements relative to the reference shown in the module manual.

A high-impedance meter on an unloaded output can show voltage that does not represent a usable signal under load. Check the connected circuit using an approved measurement method before blaming the PLC or replacing the sensor.

Can a two-wire sensor connect directly to the PLC?

Sometimes—but the pair must be electrically compatible. A two-wire electronic sensor powers itself through the load path. It can pass leakage current while OFF and retain a voltage drop while ON.

• OFF: Leakage through the input must remain within its guaranteed OFF/reset limits. The resulting input voltage must also be low enough for OFF.
• ON: After the sensor's residual voltage and cable losses, sufficient voltage must remain for the input to recognize ON.
• Load: The input must satisfy the sensor's specified operating-load range, including any minimum load current.

If the input stays ON after the target leaves, investigate leakage and the real input circuit before changing NO/NC logic. A bleed resistor or interface module can sometimes resolve a mismatch, but it needs an approved design covering resistance, dissipation, added ON current and operating extremes. Do not add a generic resistor value from another installation.

Why does the PLC miss short sensor pulses?

A correct static connection can still lose moving targets. The relevant signal is the actual pulse at the PLC input, including its shortest ON time and shortest OFF gap at maximum production speed—not just the sensor's advertised switching frequency.

Input filtering is model-specific

Some modules qualify a new state only after it remains continuous for a configured duration. For that mechanism, a short OFF gap may never reset the input even when every ON pulse is long enough. Other designs accumulate signal duration differently.

For example, Siemens documents an up/down accumulation mechanism for S7-1200 digital input filtering. Repeated pulses can affect the accumulated value. Therefore, do not apply a universal “pulse width versus filter setting” pass/fail calculation to every PLC.

Capturing an event is different from acting on it

After filtering, an event still has to reach the program through the module's update and task mechanisms. A pulse may disappear between ordinary reads. A supported latch, interrupt or high-speed counter can change that behavior, but its channel, pulse-width and configuration limits still apply.

Siemens S7-1200 pulse catch is an example: it captures after the input filter, so it cannot recover an event already rejected by that filter. It also does not provide an unlimited count of pulses between input updates. Use the appropriate counting function when every event must be counted.

Test two separate requirements: was every event captured? and did the machine respond before its process deadline? A latched event can be captured successfully while the resulting action is still too late.

What changes for 4–20 mA and 0–10 V inputs?

Analog signals need a matching analog input, not merely a spare digital terminal. Select the signal range in the hardware configuration before checking the value displayed in engineering units.

For 4–20 mA, identify who supplies the loop

Check whether the sensor is loop-powered or separately powered and whether the input supplies loop power. Follow the pair's wiring diagrams; do not combine two loop supplies inadvertently. The supply must cover the transmitter's minimum operating voltage plus the input burden and wiring drops at the required current.

Also check permitted grounding and common-mode voltage. A correctly measured current does not make an unsuitable grounding arrangement acceptable.

For voltage signals, verify the reference and scaling

For 0–10 V, identify signal return, required input impedance and permissible common-mode range. Do not assume cabinet earth is the signal reference. Then verify the channel range, raw-data format and scaling used by the program.

For an assumed linear 4–20 mA signal, 12 mA is 50% of span: (12 − 4) ÷ (20 − 4) = 0.5. It is not necessarily 50 engineering units. A 0–200 bar configuration would display 100 bar. A zero-current reading can indicate a broken or unpowered loop; distinguish the causes through diagnostics rather than treating that reading as a unique fault code.

Where should you look when the sensor and PLC disagree?

Start at the earliest point where the observation differs from what the wiring and configuration predict. Record the sensor model, input-module part number, channel, common group, filter and program tag before changing anything.

Complete the check with both target states, repeated cycles at working speed and the intended power-up sequence. Remove temporary forces and test overrides under the site's procedure. Keep the proven wiring, hardware configuration and tag mapping together so a later replacement can be checked against the same arrangement.

The practical rule: First establish the right current path, then verify electrical states, then verify timing and program interpretation. Changing software cannot repair an incompatible input circuit.