You can measure decibels with a dedicated sound level meter, a smartphone app, or professional-grade equipment, depending on how accurate you need the reading to be. A basic sound level meter costs $20 to $50 and gives readings within a couple of decibels of the true value, which is good enough for most home and workplace checks. For context, the decibel scale spans from about 25 dB (a whisper) to 140 dB (a jet engine at 100 feet), with normal conversation falling around 60 to 70 dB.
What the Decibel Scale Actually Measures
Decibels don’t work like a ruler. The scale is logarithmic, meaning each 10 dB increase represents a tenfold jump in sound intensity. A 70 dB sound isn’t twice as loud as a 35 dB sound; it’s millions of times more intense. This logarithmic design mirrors how your ears actually perceive loudness: it takes roughly 10 times the physical intensity for a sound to register as “twice as loud” to you.
The formula behind the number is straightforward in concept. You compare the intensity of the sound you’re measuring against a fixed reference point, which is the quietest sound a healthy human ear can detect at 1,000 Hz. The calculation is 10 times the logarithm of that ratio. This compression is what lets us describe sounds ranging from near-silence to physical pain using a tidy 0-to-140 scale instead of numbers stretching into the trillions.
Smartphone Apps vs. Dedicated Meters
Phone apps are the easiest way to start, but their accuracy varies wildly. A study from the National Institute for Occupational Safety and Health tested 10 popular iOS sound measurement apps against a calibrated reference microphone. The best-performing apps came within about 0.5 to 2 dBA of the true value. The worst missed by more than 13 dBA, which is a huge error when you consider that 3 dB represents a doubling of sound energy.
The main problem is hardware. Phone microphones are designed for voice calls, not precision measurement. They clip at high volumes, struggle with very low frequencies, and vary between phone models. If you use an app, look for one that has been independently tested and, ideally, allows external microphone input. For casual use (checking whether your apartment is unusually noisy, getting a rough sense of a concert’s volume), a well-reviewed app on a recent phone gets you in the right ballpark.
A dedicated Class 2 sound level meter, the type you’d find at a hardware store or online for $30 to $200, uses a calibrated microphone and gives reliable readings within about 1.4 dB. Class 1 meters, used in professional acoustics and regulatory enforcement, tighten that to within 0.7 dB but cost $500 and up. For home use, noise complaints, or checking whether your workshop needs hearing protection, a Class 2 meter is more than adequate.
Settings That Affect Your Reading
Two settings on any decent meter or app change what your reading actually means: frequency weighting and time weighting. Getting these wrong can throw off your measurement by 10 dB or more, even with perfect equipment.
Frequency Weighting (dBA vs. dBC)
A-weighting, shown as dBA, filters the sound to match how human ears hear. Your ears are less sensitive to very low and very high frequencies, and the A-weighting curve accounts for that. Nearly all noise regulations, workplace standards, and environmental guidelines are written in dBA, so this is the default setting for most measurements.
C-weighting, or dBC, treats low-frequency sounds more equally. It’s useful when you’re specifically concerned about bass, like a thumping subwoofer through a wall or low-frequency industrial hum. If you’re not sure which to use, choose A-weighting.
Time Weighting (Fast vs. Slow)
Fast response samples the sound every 125 milliseconds, capturing quick fluctuations. This is the standard setting for most noise measurements and gives you a sense of the range you’re dealing with, such as “58 to 64 dBA.” Slow response averages over a full second, smoothing out the fluctuations to give you a better picture of the overall average in a constantly changing environment. For general-purpose measurements, fast response is the standard choice.
How to Take an Accurate Reading
Distance matters more than most people realize. Sound from a single source drops off following the inverse square law: double your distance, and the intensity falls by about 6 dB. A power mower that reads 107 dB at 3 feet might read 89 dB at 20 feet. Always note the distance when you record a measurement, and if you’re comparing readings, take them from the same spot each time.
Hold the meter at arm’s length with the microphone pointed toward the sound source. Your body reflects and absorbs sound, so keeping the meter away from your torso reduces interference. If you’re measuring workplace noise exposure, position the meter near the worker’s ear level.
Reflections from walls, floors, and ceilings can inflate your readings by bouncing sound energy back toward the microphone. Outdoors on open ground gives the cleanest measurement. Indoors, you’ll typically read higher than the source alone would produce, which is actually a more honest picture of what your ears experience in that room.
Background noise folds into every measurement. If the ambient level in your room is 40 dB and the sound you’re trying to measure is 42 dB, your meter will read close to 45 dB because the energies add together logarithmically, not linearly. For a clean reading, the sound you’re measuring should be at least 10 dB louder than the background. If it isn’t, you can subtract the background energy using the decibel subtraction formula, but in practice, most people simply try to measure in the quietest environment available.
Common Decibel Benchmarks
- 25 dB: A whisper in a quiet room
- 60 to 70 dB: Normal conversation at a few feet
- 85 dB: The level at which prolonged exposure starts to risk hearing damage
- 107 dB: A power mower
- 140 dB: A jet engine at 100 feet, at the threshold of pain
These benchmarks help you sanity-check your readings. If your meter says a normal conversation is 95 dB, something is off with your setup or calibration.
Noise Levels That Matter for Health
The reason most people measure decibels is to figure out whether a sound is safe or whether a noise problem is real. Two frameworks set the boundaries.
For workplaces, OSHA sets the permissible exposure limit at 90 dBA averaged over an 8-hour shift, with a 5 dB exchange rate. That exchange rate means every 5 dB increase halves the safe exposure time: at 95 dBA, the limit drops to 4 hours. NIOSH, the research agency behind occupational health recommendations, is more conservative, recommending 85 dBA as the threshold.
For residential and community noise, the World Health Organization recommends keeping outdoor noise below 53 dBA during the day to prevent health effects from road traffic, and below 45 dBA at night to protect sleep. Their earlier 1999 guidelines set 55 dBA as the outdoor limit for preventing serious annoyance. If you’re measuring noise in your home or neighborhood, these are the numbers to compare against.
Calibrating Your Equipment
A sound level meter is only as good as its last calibration. Dedicated meters come with a calibration certificate and should be checked annually, or before any measurement session where accuracy matters. Acoustic calibrators produce a known tone (typically 94 dB or 114 dB at 1,000 Hz) that you hold over the microphone to verify the reading.
For smartphone apps, you can do a rough calibration by comparing your phone’s reading against a known reference. If you have access to a calibrated meter, measure the same sound with both devices and note the offset. Some apps let you enter a correction factor to compensate. This won’t make a phone as reliable as a proper meter, but it can bring a decent app’s readings within a useful range.