Lock the Device and Operating State

Record the sample serial number or engineering label, hardware revision, firmware, battery pack, amplifier board, driver version, enclosure revision, and DSP preset. Photograph the control panel before the run. If the product has indoor, outdoor, bass, karaoke, or power-saving modes, select one and state it.

Battery-powered output can change as pack voltage falls. A buyer may therefore request measurements at full charge and at a defined lower state, or use the approved external supply when that reflects intended operation. Lights, microphone receivers, display brightness, and power-bank loads also belong in the record because they share the electrical and thermal budget.

Fix the Acoustic Geometry Before Pressing Play

Specify room or free-field conditions, microphone model and calibration status, measurement distance, height, angle, floor treatment, nearby boundaries, and background level. Mark the speaker and microphone positions so a second operator can repeat them. A one-metre on-axis result cannot be compared directly with a reading taken closer, off-axis, near a wall, or in a reverberant room.

Large trolley and column products may need more than one microphone position because driver layout changes with height and distance. If the commercial use case includes standing listeners, seated users, or a wide retail demonstration area, retain the reference position for the formal result and add separate coverage observations rather than averaging unlike conditions.

Choose the Signal and Endpoint Deliberately

Music is valuable for listening checks but difficult to reproduce unless the exact file, level, and segment are controlled. A defined noise signal, sweep, or test sequence is better for comparison. State spectral content, crest factor where relevant, playback duration, and whether the endpoint is a distortion limit, compression threshold, protection event, or a preset maximum volume.

IEC 60268-21 describes output-based acoustical measurements for active sound systems and covers small- and large-signal behavior. It does not choose a marketing claim for a buyer. The project still needs an agreed method, environment, and acceptance boundary appropriate to the product.

Observe Compression and Protection During the Run

Increase level in documented steps and record acoustic output at each step. If a larger input increase produces little additional output, the system may be compressing or limiting. Continue only within the safe test instruction; forcing a unit beyond its protection boundary can damage a sample and does not represent customer use.

After a sustained high-output segment, repeat a lower reference level. A changed response can indicate thermal compression, battery sag, altered limiter state, or a component problem. Allow and record recovery time. The speaker-sensitivity article helps explain why electrical input alone does not predict this result.

Separate Useful Output from Unacceptable Sound

A meter can rise while the product becomes harsh, rattles, buzzes, leaks air, clips, or drives a passive radiator beyond a stable range. Pair the measurement with a trained listening check and, when available, distortion and compression plots. Note the frequency band and physical location of each artifact.

Do not hide an intermittent defect by reporting only the clean channel or the strongest sample. Compare several units if the decision will set a quotation, packaging claim, retail demonstration, or purchase order requirement. The sample-comparison guide provides a broader framework for sample identity and scoring.

Report More Than One Number

The report should preserve raw files, calibration reference, setup photographs, environmental conditions, operator, test date, sample history, stepped results, sustained result, recovery result, limiter observations, and every abnormal sound. State measurement uncertainty or repeatability where the laboratory can support it.

Acoustic Output Release Record

• The exact sample, hardware, firmware, battery, drivers, enclosure, and DSP mode are identified.
• Microphone calibration, geometry, boundaries, background noise, and environment are recorded.
• Signal, source gain, volume, duration, and endpoint are reproducible.
• Short-term output, sustained behavior, compression, protection, and recovery are reviewed.
• Rattle, buzz, clipping, harshness, air leak, and mechanical excursion are acceptance items.
• Several samples are compared when the result affects an order or public claim.
• Raw data, setup photographs, operator notes, and approved wording remain in the project file.
• Any component or firmware change routes the measurement through review and retest.

References

• International Electrotechnical Commission, IEC 60268-21:2018 — acoustical output-based measurements.