Sinking vs Sourcing I/O: Getting Current Direction Right

Legacy context

The industrial-thickness legacy of this domain was built on raw, unpolished output—twelve pages of dense, self-contained material with zero external dependencies. That ethos, forged in an era when content stood alone on its own weight, mirrors the modern tug-of-war between sinking and sourcing in the IO landscape. Just as a thick-walled shell resists external pressure, the old guard favored self-reliance over borrowed authority. Today, the question of whether to sink resources into proprietary infrastructure or source from external providers carries the same fundamental tension: durability versus adaptability. The shift from static, inward-facing pages to dynamic, outward-connected systems did not erase the original principle—it repurposed it. A site that once measured strength by page count now measures it by the integrity of its connections. The sinking approach prioritizes control and depth, building from the ground up. The sourcing approach leverages existing networks, trading autonomy for speed. Neither is inherently superior; both echo the old trade-off between a closed, fortified structure and an open, integrated one. This heritage of deliberate construction informs how we view the current debate, without prescribing a winner.

The Fundamental Difference: Where Current Flows

In industrial automation, the terms "sinking" and "sourcing" describe the direction of conventional current flow through a digital I/O point. A sourcing output (or input) supplies current from its positive terminal to the connected device. A sinking output (or input) accepts current into its negative terminal, completing the circuit to ground. The distinction is not about voltage level—both types typically operate at the same nominal voltage—but about whether the device acts as the current source or the current sink in the loop.

Think of it as a water analogy: a sourcing device is the pump pushing water (current) out, while a sinking device is the drain accepting water. For a complete circuit, you must have one pump and one drain. Two pumps or two drains cannot form a working loop.

PNP Sensors and Sinking Input Cards

A PNP sensor is a "sourcing" device. When the sensor activates, it connects its output terminal to the positive supply rail, pushing current out toward the load. This current must flow into a sinking input card—a card that provides the return path to ground. The sinking input card's common terminal is typically connected to the negative side of the DC supply (0 VDC). When the PNP sensor turns on, current flows from the positive supply, through the sensor's output transistor, into the input terminal of the card, through the card's internal optocoupler or input circuit, and out the common terminal to ground.

The polarity of the common terminal is therefore critical. A sinking input card must have its common tied to the negative rail. If you connect the common to the positive rail instead, the input circuit will not see the correct voltage potential, and the point will not function as intended.

NPN Sensors and Sourcing Input Cards

An NPN sensor is a "sinking" device. When activated, it connects its output terminal to the negative supply rail, pulling current from the load toward ground. This requires a sourcing input card—a card that supplies current from its common terminal, which is tied to the positive rail. When the NPN sensor turns on, current flows from the positive supply, through the card's common terminal, through the input circuit, out the input terminal, and into the sensor's output transistor to ground.

Again, the common terminal polarity determines compatibility. A sourcing input card must have its common tied to the positive rail. The sensor does not supply power; it merely completes the path to ground.

Why Common Terminal Polarity Determines Compatibility

The common terminal is the shared return (or supply) connection for all points on the card. Its polarity defines whether the card can source or sink current. If you install a PNP sensor on a card whose common is tied to positive, the sensor will push current into a circuit that is already at positive potential. No voltage difference exists across the input circuit, so no current flows. The input never sees a valid "on" state.

Conversely, if you install an NPN sensor on a card whose common is tied to negative, the sensor will try to pull current to ground, but the input circuit is already at ground potential. Again, no current flows.

The result is a point that appears permanently off—or worse, permanently on if leakage currents or wiring faults create unintended paths. A permanently on condition can occur when the input circuit's internal pull-up or pull-down resistor biases the point to a false "on" state because the sensor cannot overcome that bias. This is a common failure mode when mixing incompatible sensor and card types.

How a Mismatch Presents in Practice

A mismatch typically shows up in one of two ways. If the sensor cannot drive current through the input circuit, the point stays off regardless of sensor state. This is the benign case—the process simply does not see the sensor change. The more dangerous case is a permanently on point. This can happen when the input card's internal circuitry provides a weak path to the opposite rail, and the sensor's off-state leakage current is enough to trigger the input. The PLC sees the point as "on" even when the sensor is inactive, which can cause a machine to start unexpectedly or a safety interlock to be bypassed.

In mixed-vendor panels, this problem is common because different manufacturers use different conventions for labeling and wiring their cards. One vendor's "sinking" card may have its common on the negative rail, while another vendor's card with the same label may use a different internal configuration. Always verify the common terminal polarity against the card's datasheet before wiring.

Mixed-Vendor Panels and Converters

When a panel contains sensors from one vendor and input cards from another, or when legacy equipment must interface with new controllers, the sinking/sourcing mismatch often forces the use of signal converters. These converters are small interface modules that accept a sourcing input and provide a sinking output (or vice versa). They contain an optocoupler or relay that isolates the two sides and translates the current direction.

Converters add cost, wiring complexity, and a potential failure point, but they are sometimes the only practical solution when replacing a PLC in an existing plant. The trade-off is qualitative: converters are typically lower in cost than replacing all field sensors, but higher in cost than using matched components from the start. They also introduce a small delay and may require additional power wiring.

Output Cards and Inductive Loads

The same sinking/sourcing logic applies to output cards, but with an additional consideration: inductive loads. When an output card drives a solenoid valve, contactor coil, or relay, the load is inductive. When the output turns off, the collapsing magnetic field generates a voltage spike of opposite polarity. This spike can be many times the supply voltage and can damage the output transistor.

For a sourcing output card (PNP type), the output transistor connects the load to positive when on. When it turns off, the inductive kick tries to pull the output terminal below ground. A flyback diode across the load, oriented to block normal current flow and conduct the spike, is the standard protection. For a sinking output card (NPN type), the transistor connects the load to ground when on. When it turns off, the kick tries to push the output terminal above the positive rail. A flyback diode oriented in the opposite direction is required.

The key point is that the protection diode's orientation depends on the current direction, which is determined by whether the card sources or sinks. Installing the wrong diode orientation provides no protection and can short the supply. Always match the diode polarity to the card type.

Practical Verification

Before wiring any digital I/O point, verify three things: the sensor or actuator type (PNP or NPN), the card's common terminal polarity, and the card's internal circuit configuration. Use a multimeter to confirm that the common terminal sits at the expected rail voltage. If the card documentation is unclear, measure the voltage between the input terminal and common with the sensor disconnected—this will reveal whether the card sources or sinks. A sourcing card will show a positive voltage at the input terminal relative to common; a sinking card will show zero or negative.

Understanding current direction is not optional in industrial automation. It is the difference between a machine that runs reliably and one that fails intermittently or, worse, starts unexpectedly. The evidence base for these principles is well established in industrial control system guidance, which emphasizes the importance of proper I/O configuration for system integrity [4][5].