A robot noise figure is not a workplace noise assessment

Keep observation windows visible — method added 15 October 2026

A noise record can describe the right task yet leave its observation period unclear. Before using it in a buying decision, ask when the observation began and ended relative to the work. A task label cannot tell you whether the record includes replenishment, a pause, a process change or the final return movement. Keep that timing question separate from the noise result itself.

NIOSH's Understand Noise Exposure page, dated 31 January 2024 and checked 15 October 2026, distinguishes an instantaneous sound-level reading from time-weighted exposure and dose. It recommends noise-map readings during typical operation and task notes when using personal dosimeters. Those points support retaining observation context. They do not prescribe RobotAtom's worksheet or establish a universal minimum observation duration.

Start with one declared task timeline and a source that explains its boundaries. Use elapsed minutes from the same starting point for every window. If the evidence contains individual readings separated by unobserved periods, preserve those distinctions; entering the first and last timestamp as one continuous window would assert coverage that the record may not support. Do not combine different runs merely because their task names match.

The local worksheet merges overlapping entered windows before counting their duration. This avoids counting the same interval twice when two records overlap. It then lists intervals without an entered window, including any leading or trailing gap. These are bookkeeping results for the supplied offsets, not evidence that nothing was measured or that a particular event occurred during a gap.

For example, in a hypothetical ten-minute task, entered windows from minute two to four and six to eight leave intervals zero to two, four to six and eight to ten without an entered window. The useful next step is to ask the evidence owner what those intervals contain and whether another record exists. Do not fill them with an assumed quiet level, extrapolate a measured level across them or treat the covered duration as a worker dose.

ISO 9612:2025, Edition 3 published in May 2025, describes occupational-noise measurement, analysis of exposure conditions and uncertainty in its public scope. Only that scope and publication metadata were reviewed on 15 October; the full standard was not accessed. A timeline union does not implement its measurement strategies, decide sampling adequacy or establish that the selected observations represent the work. Those decisions remain with a qualified occupational-noise professional using the applicable method and local requirements.

Keep the original records, the declared task duration, each window's reference and the unresolved question together. The worksheet withholds comparison for missing references or invalid offsets and exports its entered values and limitations locally. If every minute has an entered window, the result still is not a sampling pass. Review equipment states, worker movement, instruments, uncertainty and task representativeness separately through the existing condition worksheet and responsible specialist. This addition preserves the article's original publication and review dates.

11 October 2026 — check which task the measurement describes

Before using a noise record in a buying decision, put the measured work beside the intended work. A report may be sound evidence for the task observed and still leave a different task unanswered. Record what ran, where people were and what the measurement was intended to describe before asking whether its result can be reused.

Start with one concrete activity: replenishing the cell, teaching a path, inspecting output or clearing a permitted recovery state. Identify the program, speed, load and tool, then the worker group, location and movement. These are prompts for a responsible review, not a claim that RobotAtom has assessed the activity.

NIOSH's guidance recommends measuring typical operating situations with equipment running and retaining notes on worker tasks. It also explains the different roles of area sound-level measurements and personal dosimetry. That supports asking how a record relates to the work; it does not make a stationary reading a substitute for personal exposure evidence.

Keep timing and surrounding sources visible: task duration, intermittent events, simultaneous equipment, room conditions and enclosure state. If a report leaves a condition unstated, preserve the gap. Do not replace it with the buyer's preferred operating assumption or infer that a quieter robot specification covers a different process tool.

ISO 9612:2025's public scope addresses occupational exposure and requires analysis of the exposure conditions and result uncertainty. Only that public scope was checked for this update. It does not provide a complete procedure here, choose a local legal criterion or validate the proposed robot-assisted task.

The accompanying local worksheet pairs recorded conditions with intended work. It flags blanks and different typed entries, then exports the references, reviewer basis and unresolved action. This is RobotAtom's evidence-request method. Identical text cannot prove representative conditions, and a completed form cannot establish exposure or acoustic safety.

A hypothetical replenishment task beside running conveyors may differ from a robot-only product test in both surrounding sources and worker position. The useful next step is to ask the qualified occupational-noise owner which evidence covers that task and what remains to be measured. Retain the original report unchanged and record the review's limits alongside it.

The brochure figure describes emission, not exposure

A robot's declared dB(A) figure can describe machinery emission under a named test without establishing a worker's exposure in the completed application. The result may apply at specified microphone positions, with a particular cycle, payload, mounting and environment, while the workplace adds tools, process contact, peripherals, nearby machines, room reflections and time spent at changing locations.

Preserve the product figure and its conditions as manufacturer evidence. Then assess the integrated workplace and representative tasks using the applicable occupational-noise method and local requirements.

Record exactly what the declared figure means

ABB's current IRB 6730S Revision B specification declares a product noise value measured at four points, one metre beyond the maximum robot working range and 1.5 metres above the base, during a defined VDMA cycle at maximum performance and payload. ABB also says actual application noise depends on path, payload, cycle time, mounting and environment. These are manufacturer claims for the named robot and test, not an independent workplace measurement or worker dose.

• Sound-pressure or sound-power quantity, weighting, averaging or peak metric and any declared uncertainty.
• Microphone distance, height and positions plus the boundary used for the robot's working space.
• Program or test cycle, speed, payload, mounting position, operating mode and environment.
• Whether the controller, cooling equipment, tool, process and other application components were included or excluded.
• Product model, configuration, source document, revision, publication date and verification method for the declaration.

Treat the robot as one source in the complete cell

Universal Robots' current UR20 technical sheet lists a robot noise figure as a product specification. It does not claim to include the application tool, process, other equipment, room acoustics or a worker's time-weighted exposure. The figure remains a UR claim for that product, not a whole-workplace result.

• Robot arm and controller fans, brakes, gear motion and any application-dependent operating states.
• End effector, compressed-air exhaust, vacuum equipment, pumps, welding or machining process and workpiece contact.
• Conveyors, feeders, fixtures, external axes, cooling, extraction, alarms and safety or process signalling devices.
• Adjacent machines and mobile equipment that operate at the same time as the robot cell.
• Guards, enclosures and openings whose position or condition changes propagation and exposure.

People move and tasks change

ISO 11202:2010 covers machinery emission sound-pressure levels at an operator station or other specified nearby positions under defined mounting, operating and environmental conditions. ISO 9612:2025 separately covers occupational exposure, requiring observation and analysis of exposure conditions, measurement strategies and uncertainty. These standards-body scopes establish different evidence questions; their public pages do not supply a universal limit or assess a particular site.

• Normal tending, replenishment, inspection and quality tasks, including where people stand and how long they remain there.
• Teaching, setup, recovery, cleaning and maintenance, which can place people closer to different operating sources.
• Shift pattern, task rotation, simultaneous equipment states and intermittent or peak events.
• Moving workers whose exposure cannot be represented by one fixed area measurement.
• Contractors, visitors and nearby teams whose tasks cross or border the cell's acoustic environment.

Measure representative workplace conditions

NIOSH's current US guidance illustrates why level and duration belong together and recommends measuring typical operating situations, mapping noisy areas and using personal dosimeters where appropriate. Its exposure limit is a US recommendation, not the law for every RobotAtom reader and not a universal RobotAtom acceptance criterion.

• Define applicable jurisdictional criteria, responsible occupational-noise specialist, worker groups, tasks, locations and operating states before measurement.
• Map areas with the relevant equipment running and use personal exposure measurement where a fixed reading cannot represent a worker's movement and duration.
• Capture background noise, reflective surfaces, barriers, doors and enclosure states plus intermittent and peak events where relevant.
• Preserve instruments, calibration, positions, task observations, durations, results, uncertainty, exclusions, deviations and reviewer decision.
• Reassess after material program, speed, load, tool, process, enclosure, layout, adjacent-equipment, staffing or shift changes.

Specify controls and evidence without predicting the result

ISO 4871:1996, confirmed in 2024, covers declaration and verification of machinery noise-emission values. That makes a declaration a defined product claim; it does not convert it into an occupational-exposure assessment.

• Define the acceptance method and applicable criteria rather than assuming that a product figure below a familiar number passes.
• Assign source reduction, isolation, enclosure, absorption, layout, task or scheduling controls to named owners and configurations.
• Test controls in their normal state plus foreseeable open, bypassed-for-authorized-work or maintenance conditions where applicable.
• Keep declared product data, workplace measurements and personal exposure results as different evidence types.
• Record remaining uncertainty and the conditions that require specialist review or new measurement.

Keep every claim inside its evidence boundary

ISO is authoritative for the scope and status of its emission and exposure standards. NIOSH is an authoritative US occupational-health source. ABB and Universal Robots are authoritative for their own declared product specifications. None independently verifies the completed RobotAtom reader's workplace or legal compliance.

RobotAtom can record product declarations, source states, worker tasks and locations, evidence types, applicable criteria, controls, owners and required validation. It does not determine exposure, provide a universal legal limit, certify acoustic safety or establish compliance. The responsible buyer, integrator, employer and qualified occupational specialists must decide what applies and whether the evidence is sufficient.