If you can take a clear photo, you can take a useful sound measurement. This guide walks you through reliable measurements with either a phone app or a dedicated sound level meter, and it explains how to document results so they’re helpful later.
Pick your tool
Phone apps are fine for relative comparisons and quick checks. Choose an app with A/C/Z weighting and Fast/Slow time response. For safety or compliance, use a calibrated meter with a windscreen and, ideally, a 94 dB calibrator.
Choose settings before you measure
- Weighting: A for general noise; C for bass‑heavy sources.
- Time response: Fast for transients; Slow to stabilize; LAeq for time‑average.
- Range: If your meter supports ranges, pick one that keeps the needle in the middle.
Placement and technique
- Hold the mic at ear height in the listening position.
- Point the mic as recommended by the manufacturer (often toward the source).
- Outdoors, use a windscreen; even light wind can cause spikes.
- Indoors, stand back from walls to avoid reflections skewing the reading.
Log what matters
Write down the location, date, weighting, time response, distance to source, and the reading(s): Max, Min, and LAeq for at least 30 seconds. If you’re comparing changes—like closing a door—measure both states with the same setup.
Calibrate or sanity‑check
If you have a calibrator, check 94 dB before and after a session. No calibrator? Compare two apps or meters side by side. Consistency matters more than absolute perfection for many use cases.
Worked example: quiet room vs AC on
Room idle: 33–36 dB(A) Slow, LAeq 34 dB over 60 seconds. AC on: 41–44 dB(A), LAeq 42 dB. The AC adds roughly 8 dB; perceived loudness about doubles. The record now helps you decide whether simple fixes (filters, isolation pads) made a difference later.
Common mistakes
- Using the wrong weighting for the source.
- Measuring too close to a reflective surface.
- Capturing too short a sample; spikes can mislead.
- Forgetting to document distance and settings.
Turning measurements into actions
With reliable numbers, you can evaluate changes—moving a noisy device further away, adding a soft barrier, or scheduling loud tasks when people aren’t nearby. The meter is a decision tool, not just a scorekeeper.
Settings Cheat Sheet: What to Use for Each Scenario
| Scenario | Weighting | Time Response | Metric to Record | Placement Note |
|---|---|---|---|---|
| Background noise / ambient survey | dB(A) | Slow | LAeq ≥ 60 sec | Ear height, away from walls |
| HVAC / machinery assessment | dB(A) and dB(C) | Slow | LAeq + Max | 1 m from source, note distance |
| Impact noise (tools, doors) | dB(A) | Fast or Peak | Lmax | At operator's ear position |
| Occupational exposure assessment | dB(A) | Slow | LAeq over full task duration | Lapel mic or dosimeter if available |
| Music / entertainment venue | dB(C) or dB(A) | Fast | LAeq + Lmax | Audience position, multiple spots |
| Outdoor / community noise | dB(A) | Slow | L10, L50, LAeq | 1.2–1.5 m height, windscreen essential |
| Checking if earplugs are working | dB(A) | Slow | With and without plugs | Same position, same source |
Common Measurement Mistakes and How to Fix Them
| Mistake | What Goes Wrong | Fix |
|---|---|---|
| Standing too close to a wall or corner | Reflections add 3–6 dB to readings | Stand at least 1 m from large surfaces |
| Wind noise on outdoor mic | Spikes of 10–20 dB that aren't real sound | Use a foam windscreen; measure on calm days |
| Sample too short (under 10 seconds) | Single spikes dominate; misses averages | Use LAeq over 60+ seconds for stable readings |
| Wrong weighting for the source | dB(A) reads 10+ dB lower than dB(C) for bass-heavy sources | Use dB(C) for subwoofers, HVAC, machinery |
| Not noting distance from source | Numbers can't be compared or reproduced | Always log distance in your notes |
| Comparing readings from different apps | App-to-app variation up to ±5 dB | Use the same app/device for all comparisons |
Frequently Asked Questions
How accurate are phone apps for measuring decibels?
Consumer phone apps typically have an accuracy of ±2–5 dB for relative comparisons, which is useful for identifying loud sources, comparing before/after changes, and general awareness. They're not suitable for regulatory compliance, legal noise disputes, or OSHA documentation — those require a Type 1 or Type 2 certified sound level meter with calibration records. The main limitations of phone apps are the microphone's limited dynamic range, the lack of calibration traceability, and variability between phone models. For informal use, apps like NIOSH SLM (free, iOS) or Sound Meter (Android) are reliable enough.
What's LAeq and why is it better than just reading the max level?
LAeq (Equivalent Continuous A-weighted Sound Level) is the time-averaged energy of a sound over a measurement period. It accounts for both level and duration — a source that's intermittently loud contributes less to LAeq than one that's continuously loud at the same peak. LAeq is what occupational health standards (NIOSH, OSHA) use for exposure calculations because a single high reading doesn't tell you how much total acoustic energy you were exposed to. Use LAeq for any measurement lasting more than a few seconds.
How do I document measurements for a noise complaint?
For noise complaints, document: exact GPS location and address, date and time, weather conditions (wind affects outdoor readings), microphone height and distance from the source, weighting and time response settings, LAeq over at least 60 seconds, max and min readings, and a description of the source. Take measurements in 3–5 separate sessions at different times of day. Video recording the meter display simultaneously with the source is useful for informal disputes. For formal legal or regulatory complaints, hire a certified acoustical consultant.
Should I use Fast or Slow time response on my meter?
Use Slow response for most measurements — it averages over 1 second and gives a stable reading that's easier to read and record. Use Fast response (125ms averaging) when you need to capture transient events like a door slam, an impact tool, or a brief noise event you're trying to characterize. For exposure assessment and background noise surveys, always use LAeq or Leq mode rather than relying on the instantaneous Fast or Slow display.