A safety sensor response time is not a robot-cell separation distance

The number answers only one part of the question

A light curtain, laser scanner or other safety sensor can have a documented response time without establishing where the safeguard for a robot cell should be positioned. Hazardous motion can continue while the sensor detects an approach, safety logic and communications react, the robot initiates its safety function and the complete machine comes to the required stopped state.

The buyer should ask for the full, configuration-specific evidence chain. A fast sensor is one input; it is not a cell-level separation-distance result and it does not establish that people cannot reach the hazard before the hazardous motion has stopped.

Start with the safeguard problem, not the arithmetic

ISO 13855:2024 covers the positioning and dimensioning of safeguards in relation to a person's approach to machinery hazards. Its public scope includes electro-sensitive protective equipment, pressure-sensitive devices and interlocking guards, and it states that separation distances derived from the document do not apply to safeguards used solely for presence sensing. That scope is a reason to define the actual protective function before selecting a number; it is not a distance calculation for a particular cell.

• Name the applicable jurisdiction, standards edition, risk assessment and responsible safety owner.
• Identify the hazard and every hazardous motion, including the tool, workpiece, external axes and connected machinery.
• Record who can approach, from which direction and whether a person can reach over, under, around or through the protective field.
• Identify the safeguard type, model, detection capability, orientation, mounting and intended function.
• Separate entry detection used to initiate a stop from presence sensing and other protective functions.

Record the whole response path

Pilz's supplier explanation of ISO 13855:2024 describes overall response time as including safety-related device response, the time until the machine stops and a tolerance factor. This is useful supplier guidance about the evidence categories; it is not an independent verification of a robot cell and does not replace the applicable standard or an engineering assessment.

• Sensor detection and output response under the selected configuration.
• Safety logic, input filtering, communications, safe network or discrete-I/O path and output response.
• Robot safety-function reaction and stopping performance.
• Tool, workpiece, external-axis and other machine motion that can remain hazardous.
• Tolerance, uncertainty and application-specific additions used by the responsible engineer.

Stopping performance belongs to a configuration

ABB's SafeMove Revision T manual says a stopping-supervision reaction time must be added to the applicable manipulator stopping time and lists different response paths for discrete external stops and safe-bus stops. Universal Robots publishes stopping data tied to stated joints, payload fractions, trajectories and stop categories, and says configured stopping limits can dynamically reduce robot speed. These are first-party claims for identified ABB and UR configurations—not generic values for other robots and not proof of a completed cell assessment.

• Exact robot, controller, software and safety configuration.
• Tool, workpiece, payload, centre of gravity and external axes.
• Speed, pose, trajectory, operating mode and stop category.
• Worst relevant approach to the hazard and stopping travel of all hazardous parts.
• Source document, revision, test conditions and any excluded operating state.

Geometry can still change the result

Elapsed time is not the only input. The responsible designer also has to consider the relevant approach, protective-field geometry, detection capability, reach through or around the safeguard, the location and motion of the hazard, and the additions or tolerances required for the application.

ISO 13855:2024 notes in its public abstract that different approaches can produce speeds different from its defined values. Pilz says its interpretation of the 2024 revision includes moving hazards and dynamic approach. Those statements support recording the geometry and motion assumptions; they do not justify applying a single web example or sensor-catalogue figure to every installation.

Require an auditable evidence package

ISO 10218-2:2025 addresses the integration, commissioning, operation and maintenance of industrial robot applications and cells. That application-level scope supports treating the safeguard decision as an integration record rather than a component shopping comparison. The purchased, jurisdictionally applicable standards and qualified safety specialists remain necessary for the actual design and validation.

• A controlled calculation or design record naming the hazard, safeguard, standards basis, inputs, assumptions and author.
• Traceable component and robot data for the exact installed versions and settings.
• A method for measuring or otherwise validating the complete installed stopping behaviour under the relevant configurations.
• Results, uncertainty treatment, deviations, reviewer, decision and retained raw evidence.
• Retest triggers for payload, tool, speed, path, logic, firmware, network, safeguard, layout or connected-machine changes.

Use unknowns as gates, not assumptions

If the worst relevant stopping performance, signal path, geometry or approach assumption is unknown, the claim is not ready to pass a safety evidence gate. Route it to measurement, engineering analysis or specialist review instead of substituting a typical sensor value, brochure stop time or simulation result.

RobotAtom can preserve each source, configuration, assumption, owner, date, result and change trigger. It does not calculate or approve separation distance, certify a safeguard or robot cell, decide that a risk assessment is complete, or establish legal compliance. Those decisions belong to the responsible integrator, user and qualified safety professionals for the actual application.