Sensor Timer Functions: When Local Delay Helps PLC Logic
Built-in sensor timers can reject chatter, stretch short pulses, and simplify local timing—but they do not replace PLC diagnostics or safety logic. Learn how to apply on-delay, off-delay, and one-s...
Many smart photoelectric, capacitive, and proximity sensors can delay, extend, or pulse their switching output before the signal reaches the PLC. Used carefully, this local processing can reject chatter and make short events visible. Used casually, it can hide the real process timing and send maintenance teams searching in the wrong program.
The timer belongs at the sensor only when the behavior is part of the sensing requirement. Machine sequencing, fault response, and operator-visible timing usually remain easier to govern in the controller.
What the common timer modes actually do
On-delay filters short detections
An on-delay requires the sensing condition to remain active for the configured time before the output changes state. A shorter event is rejected. This can suppress vibration, splash, or momentary background reflections, but it also adds latency to every valid detection.

The output changes only after the input remains active for the Ton interval. Timing diagram provided by Carlo Gavazzi and retained on the PLC ProTech CDN.
Off-delay holds the output after detection ends
An off-delay changes the output immediately when the target appears, then keeps it active for a defined interval after the target disappears. It can bridge small gaps between packages or maintain a signal long enough for a downstream device to respond.
Combined delay and one-shot modes reshape events
An on/off-delay applies independent qualification to both transitions. A leading-edge one-shot produces a fixed-duration pulse when detection begins; a trailing-edge one-shot produces it when detection ends. The pulse width is configured locally rather than determined by target length.
Carlo Gavazzi’s IO-Link sensor documentation lists on-delay, off-delay, combined delay, and leading- or trailing-edge one-shot modes. Other vendors may use different names or apply timer logic before or after output inversion, so the product manual is the final authority.
Local timing solves three practical problems
First, it can debounce a mechanically unstable or optically noisy event close to the source. Second, it can stretch a pulse that is shorter than the PLC input filter, network update, or task interval. Third, it can provide the same response to a local hardwired output and an IO-Link process-data channel when the device supports both.
The benefit is deterministic behavior at the sensing point, not “free redundancy.” A timed sensor output does not keep a coordinated machine safe or productive after a PLC failure. It simply continues to apply its configured signal rule while powered.
Why PLC timing is often the better home
A controller timer is visible in the program, can be alarmed, can be changed under access control, and can use mode, recipe, and machine-state context. Engineers can see the raw input and the qualified signal together. That makes root-cause analysis easier when product spacing, speed, or reflectivity changes.
Sensor-side timing hides the raw event from a discrete PLC input unless a second channel or IO-Link diagnostic value exposes it. If the machine misses parts, the controller may only see the reshaped signal and cannot distinguish a weak target from an aggressive delay setting.
The scan-cycle issue is similar to one-shot logic. The guide to one-shot pulses in Studio 5000 explains why event width and task execution must be evaluated together rather than assumed.
Calculate timing as an end-to-end budget
The response seen by the machine includes sensor response time, the configured timer, input-module filtering, IO-Link cycle time or network update, controller task period, program execution, output delay, and actuator response. Worst-case latency matters more than a single catalog number.
For pulse stretching, set the output duration with margin above the slowest legitimate acquisition path. For on-delay filtering, verify that the shortest valid target remains present longer than the delay at maximum line speed. A setting that works during a slow commissioning run can reject good product at production speed.
IO-Link makes parameters manageable—and easier to change accidentally
IO-Link can expose timer mode and value as parameters, enabling recipe download and device replacement. It also creates governance requirements. Store the approved parameter set, control who can write it, and record units because some devices allow milliseconds, seconds, or minutes.
During replacement, confirm the device identity and IODD version before restoring parameters. The IO-Link commissioning guide covers port modes, process data, IODD handling, and validation.
Do not use a standard sensor timer as a safety function
A standard timed output is not a safety-rated delay, muting function, or protective stop. Safety applications require suitable devices, architecture, diagnostics, and validation. If the timed signal influences a hazardous motion, the risk assessment must determine whether the function belongs in a safety controller or certified device.
A commissioning test that exposes the hidden edge cases
Record the timer mode, value, units, output inversion, and whether the setting applies to the hardwired output, IO-Link channel, or both. Test pulses just below and just above the threshold. Change line speed, introduce realistic gaps, cycle power, disconnect IO-Link, and verify the fallback state.
Trend the raw detection value when available, the timed switching signal, the PLC-qualified bit, and the final machine action. This evidence shows where latency was added and prevents duplicate filtering in both the sensor and PLC.
Editorial view: keep local intelligence local in scope
PLC ProTech recommends sensor timers for signal conditioning that is inseparable from the sensing task: chatter rejection, gap bridging, and pulse stretching. Sequence timing and fault decisions should remain in the controller unless a documented architecture requires otherwise.
A smart sensor should reduce ambiguity at the input. If its timer makes the raw process invisible, it has traded a few ladder instructions for a harder maintenance problem.
FAQ
What is the difference between on-delay and off-delay in a sensor?
On-delay waits before turning the output on and rejects inputs shorter than the delay. Off-delay turns on immediately but waits before turning off after detection ends.
Can a sensor one-shot capture an event that the PLC would miss?
It can stretch the output pulse so the PLC input path has time to detect it. The pulse must still exceed the worst-case input filter, network update, and task interval with suitable margin.
Should the same signal be filtered in both the sensor and PLC?
Only with a documented reason. Stacked filters add latency and can hide valid events. Keep one stage visible and record the total timing budget.
Do IO-Link timer settings survive sensor replacement?
They can when the master supports parameter storage and the replacement device is compatible. Commissioning must verify device identity, IODD revision, units, and the restored values.