Detecting Unplugged Safety Sensors in PLC Control Systems
Detect unplugged safety sensors in PLC systems using NC sensors, diagnostic input modules like Allen-Bradley 1756-IA8D, or dual-contact sensor configurations...
On many discrete machines, a single proximity or light-curtain channel is treated as a safety or quality interlock even though the input is a conventional discrete point. When that channel uses a normally open (N.O.) sensor, the controller cannot tell the difference between a clear field and a removed connector. Both conditions present FALSE to the ladder. Operators who discover the shortcut can park a guard open, unplug a tooling sensor, or bypass a presence check without generating an alarm. The machine continues to run under the false assumption that the process is healthy.
Presence and tooling sensors on a machine cell must remain electrically visible when the cable is removed, not only when the target is absent.
This article explains why N.O. sensing hides open-wire faults, how normally closed (N.C.) and dual-contact devices restore visibility, and where Allen-Bradley diagnostic input modules such as the 1756-IA8D and 1756-IB16D fit into a ControlLogix architecture. It also draws a hard line between PLC diagnostics and certified safety architectures that rely on safety relays or safety PLCs for SIL 2 / PLd functions. Publication 1756-TD002 remains the primary reference for ControlLogix digital I/O electrical characteristics.
Why normally open sensors hide unplugged cables
A discrete input that expects a sourcing or sinking sensor evaluates current through the channel. With an N.O. device, current flows only when the target is detected. When the target is absent, the channel is open and the input bit is FALSE. When the cable is unplugged, the channel is also open and the input bit is again FALSE. The PLC application therefore has no independent state that means “sensor missing.” Quality logic that simply waits for TRUE before advancing a cycle will treat both conditions as “clear to proceed,” which is exactly the wrong outcome for guarding and presence applications.
| Condition | N.O. sensor | N.C. sensor |
|---|---|---|
| Connected, target absent | FALSE | TRUE |
| Connected, target present | TRUE | FALSE |
| Cable removed / open wire | FALSE | FALSE |
With an N.C. sensor, the healthy idle state is TRUE. Detection of the target opens the contact and drives the bit FALSE. An unplugged cable also drives the bit FALSE, but that state only appears when something is wrong relative to the expected idle TRUE. Application logic can therefore require TRUE before enabling motion, and treat a sustained FALSE without a corresponding process reason as a fault. The distinction is not semantic; it is the difference between a silent bypass and a forced stop.
Dual-contact sensors and hybrid architectures
Many industrial proximity and guard switches offer dual-contact packages: one N.O. path for machine logic and one N.C. path for monitoring or safety string participation. A practical pattern routes the N.O. contact into a standard PLC input used for sequence or quality decisions, while the N.C. contact feeds a safety relay or safety I/O channel that must remain closed for the machine to enable. If the connector is removed, the N.C. path opens and the safety relay drops regardless of what the PLC thinks. If the target is simply present, the N.O. path changes while the N.C. monitoring path can still prove continuity of the cable and device body.
Where dual-contact hardware is not available, series wiring of an N.C. “device present” loop through a dedicated monitoring input still helps, provided the installer understands that a single wire break anywhere in the loop will look like a missing device. Labeling, junction-box documentation, and periodic function tests remain mandatory. Do not rely on a lone PLC input bit as the only layer of protection for personnel hazards.
Diagnostic modules: 1756-IA8D and 1756-IB16D
ControlLogix diagnostic digital modules add open-wire and short detection that standard modules lack. The 1756-IA8D AC diagnostic input module can indicate an open-wire condition when the field circuit draws less than approximately 2 mA with excitation present. The 1756-IB16D DC diagnostic input module flags open-wire conditions below roughly 0.5 mA. These thresholds matter because some sensors and legacy field devices leak enough current to keep a non-diagnostic module “happy” while still sitting near the edge of a true open. Rockwell publication 1756-TD002 tabulates the electrical limits and diagnostic behavior; always design to the published figures for the specific catalog number and series.
Diagnostic ControlLogix input modules report open-wire states that standard discrete cards cannot distinguish from a legitimate OFF.
Legacy Genius I/O systems are often cited in retrofit discussions because their leakage behavior around 2 mA influenced how older AC circuits were designed. When migrating Genius points into ControlLogix, do not assume that the same sensor wiring will automatically produce clean diagnostic status on a 1756-IA8D. Measure quiescent current, confirm sensor off-state leakage, and adjust pull-up or excitation networks if the diagnostic bit chatters or never asserts. Diagnostic status bits should be mapped into alarming and inhibit logic, not ignored as optional status.
Implementation procedure
- Inventory every sensor used as a guard, presence, or tooling interlock. Mark polarity (N.O. versus N.C.) and whether the connector can be reached by operators.
- Prefer N.C. or dual-contact devices for any channel that must prove cable integrity. Update electrical drawings and PLC tags to match the new idle state.
- Where solid-state channels remain, specify diagnostic modules (1756-IB16D for DC, 1756-IA8D for AC) and commission open-wire tests by unplugging each channel under controlled conditions.
- Map diagnostic fault bits to first-out alarming and inhibit permissive logic. Require a verified reset after an open-wire event.
- For SIL 2 / PLd functions, implement the safety function in safety relays or a safety PLC architecture. Never claim a standard PLC program alone as the safety integrity measure.
Prove open-wire detection on every channel
Verification is straightforward and should be recorded in the commissioning checklist. With power applied and the machine inhibited, unplug each critical sensor and confirm that the expected alarm or safety drop occurs within the designed response time. Reconnect and confirm that the idle N.C. state returns TRUE before clearing the fault. Pitfalls include leaving N.O. sensors on “safety” tags out of habit, using diagnostic modules without wiring excitation correctly, and assuming that a Genius-era leakage budget still applies after a platform change. Another common error is treating a diagnostic open-wire bit as a soft warning while the machine continues to cycle.
Finally, remember that diagnostic discrete I/O improves maintainability and reduces silent bypasses, but it does not by itself create a certified safety function. Pair the PLC diagnostics with hardware safety products sized to the risk assessment, and keep spare PLC and I/O modules aligned with the same ControlLogix platform standards used on the rest of the cell.
About the Author
Mark Townsend | Senior Automation Engineer – Allen-Bradley Systems
Mark Townsend is a senior automation engineer with more than 18 years on Allen-Bradley platforms spanning ControlLogix, CompactLogix, and legacy SLC-500. His day-to-day work is RSLogix / Studio 5000 logic and FactoryTalk View HMI bring-up on aging and mixed fleets.