Skip to content
The Outlook server is temporarily down. Please send your requests to us on WhatsApp.
Stock is confirmed when you order, and we will advise you as soon as your order is confirmed. Need help with a part? Email orderdesk@kkmsolutions.com.
Ships worldwide — lead time quoted per part orderdesk@kkmsolutions.com
KKM Industrial and Automation Solutions Industrial Automation & MRO Parts Need a part? Email the order desk orderdesk@kkmsolutions.com
Email the order desk Get a quote
PLC, HMI & Industrial Networking

PLC I/O Signal Types: A Wiring Reference

KKM Solutions · August 11, 2026

A practical wiring reference for PLC digital, analog, and temperature I/O: sinking versus sourcing, leakage current bleed resistors, 4-20mA loops, and RTD.

PLC I/O Signal Types: A Wiring Reference — KKM Solutions technical article

Short answer

  • A PLC input fails to see a signal, or an analog loop reads erratically, when the field device's output type does not match the card's input circuit, such as an NPN sensor on a sinking input or a PNP sensor on a sourcing input.
  • Two-wire sensors add leakage current that can read as a false on signal, which a bleed resistor wired in parallel with the input diverts, while 4-20mA current loops resist voltage drop and detect a broken wire as 0mA.
  • For temperature, 4-wire RTDs eliminate lead resistance error and thermocouples need cold-junction compensation to read correctly.

When a digital input card fails to see a sensor signal, or an analog loop reads erratically, the root cause is almost always a mismatch between the field device's output type and the card's input circuit. This reference covers the physical and electrical mechanisms behind PLC I/O signal types so you can specify and wire them correctly the first time. All electrical work described here requires a qualified technician and adherence to lockout/tagout procedures.

How do sinking, sourcing, and leakage affect digital inputs?

\n

Digital input (DI) circuits are fundamentally about completing a circuit to a common reference. The terms "sinking" and "sourcing" describe the direction of conventional current flow through the card. A sinking input card provides a path to ground (or the negative DC bus). Current flows from the positive supply, through the sensor, and into the input terminal. A sourcing input card supplies a positive voltage to the field device. Current flows from the card's input terminal, through the sensor, and to the negative common.

This must match the sensor's output transistor. An NPN sensor is a sinking device; it sinks current from the load to ground. A PNP sensor is a sourcing device; it sources current from the positive supply to the load. Wiring an NPN sensor to a sinking input leaves no potential difference to drive current. Wiring a PNP sensor to a sourcing input does the same. In both cases, the input never turns on. To resolve this, you either change the sensor type, change the card, or use interface relays to invert the signal.

Three-wire sensors (NPN or PNP) are straightforward, but two-wire devices—like proximity switches or mechanical limit switches with LED indicators—present a leakage current problem. A two-wire sensor must draw a small continuous current to power its internal electronics even when it is in the "off" state. This leakage current flows through the input circuit. If the leakage current exceeds the input card's "off" threshold current, the PLC will read a false "on" signal.

The standard fix is a bleed resistor (or dummy load) wired in parallel with the input terminal and the common. The resistor diverts the leakage current away from the input circuit. Calculate the resistor value based on the sensor's maximum leakage current and the card's off-state threshold voltage. You must also size the resistor's wattage for continuous power dissipation using P = V²/R, as it will be constantly energized. Selecting the right sensor terminal blocks at the junction can simplify wiring these resistors and managing the dense field wiring typically found at the machine interface.

How do relay, transistor, and triac outputs differ?

Digital output (DO) cards translate PLC logic into physical switching. The three primary output types—relay, transistor, and triac—have distinct physical mechanisms, speeds, and lifespans.

  • Relay Outputs: These use mechanical dry contacts. They are highly flexible, capable of switching AC or DC loads at higher currents (typically up to 2A or 5A per point). However, they are slow. Mechanical relays have operate times in the tens of milliseconds, making them unsuitable for high-speed pulse outputs. They also have a finite mechanical life, eventually failing due to contact wear or mechanical fatigue. Switching inductive loads without proper suppression generates arcs that pit the contacts, drastically shortening their life.
  • Transistor Outputs: These use solid-state switches (typically NPN sinking or PNP sourcing). They are fast, capable of switching at kilohertz frequencies for stepper or servo pulse trains. They are strictly DC. Polarity matters; reversing the wiring can destroy the output transistor. They have no moving parts and therefore no mechanical wear, but they are sensitive to overcurrent and short circuits. Fast-switching inductive loads require flyback diodes to clamp voltage spikes.
  • Triac Outputs: These are solid-state switches designed for AC loads only. They turn on when a gate receives a trigger signal and turn off when the AC waveform crosses zero. Like transistors, they are susceptible to overvoltage and overcurrent, requiring proper fusing and often a metal-oxide varistor (MOV) for transient suppression.

When specifying a system, match the output type to the load. Use relays for general-purpose interposing and AC loads, transistors for high-speed DC control, and triacs for solid-state AC switching. The choice of output card is dictated by the field devices, but the architecture is ultimately controlled by the PLC CPUs and systems executing the logic.

Should an analog signal use a current loop or voltage?

Analog I/O transmits continuous process variables. The two dominant standards are 4-20mA current loops and 0-10V voltage signals.

The 4-20mA current loop is the industrial standard for noise immunity and distance. The transmitter regulates the current in the loop between 4mA and 20mA, representing 0% to 100% of the process range. Because the signal is a current, it is immune to voltage drops caused by wire resistance. The loop can span hundreds of meters without signal degradation. The 4mA "live zero" provides a distinct advantage: if the wire breaks, the current drops to 0mA. The PLC can detect this as a fault condition, distinguishing it from a valid 0% process reading. Two-wire transmitters draw their operating power from the loop itself, requiring a DC power supply in series with the transmitter and the PLC input.

By contrast, a 0-10V signal is simpler but susceptible to voltage drop over long cable runs and induced noise from nearby drives or motors. A broken wire results in a 0V reading, which is indistinguishable from a valid 0% signal unless additional logic or hardware is used. Voltage signals are best suited for short-distance runs inside a control panel.

When electrical noise or ground loops corrupt analog signals, installing signal conditioners and isolators between the field device and the PLC input can break ground loops, amplify weak signals, and convert voltage to current or vice versa.

What do ADC bits mean in engineering units?

Analog-to-digital converters (ADCs) in analog input cards sample the continuous signal and convert it to a digital integer. The resolution of the ADC determines how many discrete steps are available. A 12-bit ADC provides 2¹², or 4096, counts. A 16-bit ADC provides 2¹⁶, or 65536, counts.

To understand what this means in engineering units, calculate the resolution per count. For a 0-10V signal on a 12-bit card, the resolution is 10V / 4096 = 0.00244V, or 2.44mV per count. If the sensor measures pressure from 0 to 100 PSI, the theoretical resolution is 100 PSI / 4096 = 0.0244 PSI per count.

In practice, the effective resolution is lower than the theoretical. Input cards have inherent noise and calibration errors. A 12-bit card might only have 10 or 11 bits of usable, stable data. If your process requires detecting a change smaller than the resolution per count, you must specify a card with higher bit resolution. However, higher resolution does not improve sensor accuracy; it only allows the PLC to see finer increments of the sensor's signal.

How do RTDs and thermocouples connect to temperature inputs?

Temperature measurement requires specific input cards designed for the low-voltage or resistance characteristics of the sensors.

Resistance Temperature Detectors (RTDs) measure temperature by correlating the resistance of a metal (typically platinum, as in a Pt100) to temperature. Because the RTD is a resistor, the PLC must supply a small excitation current and measure the resulting voltage drop. The lead wires connecting the RTD to the card have their own resistance, which adds to the measured value and creates an error.

A 2-wire RTD cannot compensate for lead resistance and is only suitable for short runs where the error is negligible. A 3-wire RTD adds a third lead. The card measures the resistance of the lead wires and subtracts it from the total, assuming all three leads have identical resistance. This is the industrial standard but is vulnerable to errors if the leads are different temperatures or gauges. A 4-wire RTD uses two leads for excitation and two for sensing the voltage drop. Because the sensing circuit has high impedance, no current flows in the sense leads, eliminating lead resistance error entirely. For high-precision applications, 4-wire is mandatory.

Thermocouples generate a small millivolt signal proportional to the temperature difference between the measuring junction (the hot end) and the reference junction (where the thermocouple wire connects to copper). The PLC card must know the temperature at the reference junction to calculate the absolute temperature at the measuring junction. This is cold-junction compensation (CJC). The card uses a thermistor or integrated circuit at its terminal block to measure the local temperature and mathematically correct the reading. If the CJC sensor fails or is bypassed, the thermocouple reading will drift significantly with ambient temperature changes inside the panel.

Expanding I/O for these specific signal types often requires adding remote racks or distributed architecture. Using I/O modules and remote I/O allows you to place the analog or temperature cards physically close to the process, minimizing long analog cable runs and reducing noise pickup.

If you need to source specific I/O cards, interface relays, or signal conditioners to resolve a wiring mismatch, KKM Industrial can cross-reference your requirements and source the necessary components for your automation architecture.

Related reading: what a PLC is and how it works; how current transformers and transducers work.

Worth a look next: The 4-20 mA Signal Explained: Why Industry Uses It; Modbus RTU vs Modbus TCP: What Is the Difference?.

Worth a look next: The PLC Scan Cycle Explained.

Need a quote on a specific part?

Our order desk replies within 1 business day. No account needed. Or email orderdesk@kkmsolutions.com.

Request a Quote

Frequently asked questions

Why does a PLC digital input never turn on with an NPN sensor?

An NPN sensor is a sinking device, so wiring it to a sinking input leaves no potential difference to drive current, and the input never turns on. A PNP sensor on a sourcing input fails the same way. The fix is to change the sensor type, change the card, or use interface relays to invert the signal.

How do you fix false on signals from a two-wire sensor?

A two-wire sensor draws a small continuous current to power its electronics even when off, and that leakage current can exceed the card's off threshold and read as a false on. A bleed resistor wired in parallel with the input terminal and common diverts the leakage current. Its wattage must suit continuous power dissipation because it is constantly energized.

Why is a 4-20mA loop better than 0-10V for long cable runs?

A 4-20mA loop is a current signal, so wire resistance voltage drops do not affect it, and the loop can span hundreds of meters without signal degradation. The 4mA live zero means a broken wire drops the current to 0mA, which the PLC detects as a fault. A 0-10V signal reads 0V on a broken wire, indistinguishable from a valid 0% signal.

How does ADC bit resolution translate into engineering units?

A 12-bit ADC provides 4096 counts, so a 0-10V signal gives 2.44mV per count and a 0 to 100 PSI sensor gives 0.0244 PSI per count. Effective resolution is lower because of noise and calibration errors, so a 12-bit card might have only 10 or 11 bits of usable data. Higher resolution does not improve sensor accuracy.