Measuring noise in dBA requires a device with A-weighted frequency filtering, proper positioning relative to the sound source, and calibration before and after each session. You can use a dedicated sound level meter for professional accuracy or a smartphone app for quick estimates, with the best apps coming within ±2 dBA of professional instruments.
What dBA Actually Measures
The “A” in dBA stands for A-weighting, a filter applied to raw sound pressure readings that mimics how human ears perceive loudness. Your ears are less sensitive to very low and very high frequencies and most sensitive to the mid-range frequencies in speech. A-weighting adjusts for this by reducing the contribution of low-frequency sounds and slightly boosting mid-range ones, roughly following the shape of equal-loudness curves at moderate volume (around 40 dB at 1 kHz).
Without A-weighting, a plain decibel reading treats all frequencies equally, which can overstate the perceived loudness of deep rumbles and understate the annoyance of mid-range noise. That’s why dBA is the standard unit in workplace regulations, environmental noise ordinances, and hearing conservation programs. Other weightings exist: C-weighting (dBC) keeps low frequencies closer to their true level, making it useful for measuring peak impacts or explosions, but for general noise assessment, dBA is what regulators and health agencies use.
Equipment Options
Dedicated Sound Level Meters
Professional sound level meters are classified under international standard IEC 61672 into two tiers. Class 1 meters have tighter accuracy tolerances and are required for regulatory compliance, environmental monitoring, and legal disputes. Class 2 meters have wider tolerances (±2 dBA) and work well for workplace screening, general surveys, and preliminary assessments. For most non-regulatory purposes, a Class 2 meter is sufficient and considerably cheaper, typically ranging from $200 to $500 compared to $1,000 or more for Class 1 instruments.
Both classes include built-in A-weighting filters, display real-time dBA readings, and can log data over time. Many also calculate Leq, the equivalent continuous sound level, which represents the steady noise level that would deliver the same total sound energy as the actual fluctuating noise over your measurement period. Leq is the metric you need for comparing against workplace exposure limits.
Smartphone Apps
NIOSH tested a range of smartphone sound measurement apps and found that three had mean differences within ±2 dBA of a reference-grade system, putting them in the same accuracy ballpark as a Class 2 meter. When paired with an external calibrated microphone, the best apps came within ±1 dB of reference measurements. Without an external mic, results vary more because phone microphones aren’t designed for acoustic measurement and differ between models. If you’re using a smartphone app for anything beyond a rough check, an external microphone dramatically improves reliability.
The NIOSH SLM app (for iOS) was developed specifically for occupational noise measurement and remains a solid free option. On Android, several apps perform well, but accuracy depends heavily on your phone’s hardware. Treat smartphone readings as useful estimates rather than legally defensible numbers.
How to Set Up a Measurement
Before you start, calibrate your meter. Acoustic calibrators generate a known sound pressure level (typically 94 dB or 114 dB) at 1,000 Hz. You place the calibrator over the microphone, compare the meter’s reading to the known value, and adjust sensitivity if there’s any drift. International standards require calibration both before and after each measurement session. If the readings differ by more than a small amount between the two checks, your data from that session may not be reliable.
Position matters. For workplace noise assessments, hold the meter at ear height of the person whose exposure you’re measuring, with the microphone pointed toward the dominant noise source. Keep the meter at arm’s length to avoid reflections off your body affecting the reading. If you’re mounting the meter on a tripod, place it where the worker’s head would normally be.
For environmental noise (traffic, construction, neighbor complaints), measure at the location where the noise is experienced, not at the source. Outdoors, stay at least a meter or two away from walls, fences, and other reflective surfaces that can artificially boost readings. Indoors, avoid corners and positions directly against walls, where reflected sound energy accumulates.
Wind creates low-frequency turbulence across the microphone that registers as noise. Use a windscreen (the foam cover that comes with most meters) for any outdoor measurement, even on days that feel calm.
Taking Accurate Readings
Most environments don’t produce a constant noise level. Traffic surges, machinery cycles, and conversations create fluctuations that make a single snapshot misleading. There are several approaches depending on what you need.
For a quick spot check, set your meter to “slow” response (which averages readings over one second) and note the range over a minute or two. This gives you a general sense of the noise environment. For a more rigorous assessment, use the Leq function if your meter has one. Leq integrates all the fluctuations over your chosen time period into a single number representing the constant level that would deliver equivalent sound energy. A five-minute Leq in a factory, for example, accounts for both the loud bursts when a press fires and the quieter intervals between cycles.
For full-shift workplace exposure, you need either a dosimeter (a small device worn on the body that accumulates exposure data over hours) or a sound level meter logging Leq values at regular intervals throughout the shift. The goal is to calculate a time-weighted average across the full work period, since hearing damage depends on both loudness and duration.
Key Noise Thresholds to Know
Your measurements are most useful when compared against established limits. NIOSH sets a recommended exposure limit of 85 dBA averaged over an eight-hour workday, using a 3 dB exchange rate. That exchange rate means every 3 dBA increase cuts the safe exposure time in half: 85 dBA is safe for 8 hours, 88 dBA for 4 hours, 91 dBA for 2 hours, and so on.
OSHA’s permissible exposure limit is higher at 90 dBA for an eight-hour day, and it uses a more lenient 5 dB exchange rate, meaning exposure time halves every 5 dBA instead of every 3. The practical difference is significant. At 95 dBA, NIOSH allows about 47 minutes of exposure while OSHA allows 4 hours. Most hearing conservation professionals recommend following the stricter NIOSH guideline.
For context, normal conversation runs about 60 to 65 dBA, a vacuum cleaner about 70 to 75 dBA, a gas lawn mower about 85 to 90 dBA, and a rock concert or chainsaw 100 to 110 dBA. If your measurement consistently reads above 85 dBA in a space where people spend extended time, hearing protection or noise reduction measures are warranted.
Common Measurement Mistakes
The most frequent error is measuring too close to a single source and assuming that reading represents the overall noise environment. Sound level drops roughly 6 dB each time you double your distance from a point source outdoors, so position matters enormously. Always measure at the location where people actually are, not at the machine itself, unless you’re specifically trying to characterize that source.
Skipping calibration is another common problem. Microphone sensitivity drifts with temperature, humidity, and age. A meter that was accurate last month could easily be off by several dB today. Since the decibel scale is logarithmic, a 3 dB error means you’re measuring half or double the actual sound energy.
Relying on a single short measurement in a variable environment also leads to bad data. A 30-second reading during a quiet interval in a factory tells you nothing about the noise workers experience when all machines are running. Measure during representative conditions, and when noise varies, measure long enough for the Leq to stabilize. You’ll know it’s stable when adding another minute of measurement barely changes the number.
Finally, forgetting to check your weighting setting before measuring is surprisingly easy. Many meters default to C-weighting or unweighted (Z) mode. If you’re comparing against any standard noise limit, confirm your meter is set to A-weighting before you press record.