Measuring a Room
A single reading in the middle of a room tells you almost nothing. A useful measurement takes twenty minutes and follows a procedure.
When acoustic measurement is part of a wider workplace review, accountability and responsibility in the workplace can add operational context, but it should not be treated as a substitute for calibrated sound data.
Before starting
Decide what question you are answering.
"Is this hazardous" needs levels at the positions people occupy, over a representative period.
"Why is this room unpleasant" needs something other than a level — reverberation and speech intelligibility, which a sound level meter does not report.
For an independent reference on this subject, see Interacoustics decibel basics.
"What changed" needs a repeat of an earlier measurement with the same method, and consistency then matters more than accuracy.
The procedure
Calibrate first. Before and after, and record both.
Measure where people are, at ear height, not in the centre of the room and not against a wall — reflections raise readings near surfaces.
Hold the instrument away from your body, at arm's length. A person is an acoustic obstacle and standing behind your own meter changes the reading.
Average over a representative period. A minimum of several minutes, and long enough to include the room's cycles — a machine that runs intermittently, a phone that rings, a door that opens.
Take several positions, not one, and note them.
And record the settings: weighting, time response, and averaging period. A level without those is not a measurement and cannot be compared to anything.
The four mistakes
A spot reading instead of an average. The instantaneous value fluctuates by many decibels and the number you write down is whichever moment you looked.
Measuring the loudest moment and calling it the level. Dose is about averages over time, and the peak is a different quantity with its own limits.
Wind, outdoors or near ventilation, which produces readings unrelated to anything audible. A windscreen is not optional outdoors.
And measuring the wrong thing entirely — a room's level when the question was somebody's exposure, which follows the person rather than the position.
When a room measurement is not the answer
When the question is a person's dose.
A dosimeter worn for the shift measures what that person received, including movement between areas and the exposures a fixed position misses.
Room measurements find where the problem is. Dosimetry finds who has it. They answer different questions and one does not substitute for the other.
What to write down
Date, time, instrument, calibration before and after, positions, settings, duration, and what was operating.
Without this the numbers are unusable in six months, including by you — and for anything contested the record is the substance rather than the figure.
The short version
- Decide the question first: hazard, unpleasantness and change need different measurements
- Calibrate before and after and record both; measure at ear height where people actually are, away from walls and your own body
- Average over a representative period that includes the room's cycles, take several positions, and record weighting, time response and duration
- Four mistakes: a spot reading instead of an average, treating the peak as the level, wind without a windscreen, and measuring position when the question was exposure
- Room measurements find where the problem is; dosimetry finds who has it
- Write down date, instrument, calibration, positions, settings, duration and what was running, or the numbers are unusable in six months