Loudness is measured in decibels (dB) for physical sound intensity, but perceived loudness uses specialized scales called phons and sones that account for how human hearing actually works. The distinction matters because our ears don’t treat all frequencies equally: a 100 Hz bass tone and a 3,000 Hz midrange tone at the same decibel level won’t sound equally loud to you. Several tools and units exist to bridge the gap between raw sound pressure and what you actually hear.
Decibels: The Basic Unit
The decibel is the most familiar unit for measuring sound level. It quantifies the physical intensity of sound pressure in the air on a logarithmic scale, meaning each step up represents a multiplication of energy rather than a simple addition. Doubling the power of a sound source adds only 3 decibels. But here’s the counterintuitive part: to perceive a sound as twice as loud, you need roughly a 10-decibel increase, which corresponds to a tenfold increase in acoustic power.
This mismatch between physical energy and perceived loudness is why decibels alone don’t fully capture how loud something sounds to you. A jackhammer at 100 dB doesn’t sound twice as loud as a vacuum cleaner at 50 dB. It sounds dramatically louder, in ways the decibel number alone doesn’t convey.
Phons and Sones: Measuring What You Hear
Because decibels only describe physical intensity, researchers developed two additional scales that factor in human perception. The phon scale is built on equal loudness contours, curves that map out which combinations of frequency and decibel level sound equally loud to a listener. The reference point is always a 1,000 Hz tone. If a sound at any frequency is perceived as loud as a 60 dB tone at 1,000 Hz, that sound has a loudness of 60 phons. At 1,000 Hz, the phon value and the decibel value are always identical.
The sone scale goes a step further by creating a linear relationship with perceived loudness. On this scale, 2 sones sounds twice as loud as 1 sone. The anchor point is 40 phons, which equals 1 sone. From there, every 10-phon increase doubles the sone value: 50 phons equals 2 sones, 60 phons equals 4 sones, 70 phons equals 8, and so on up to 100 phons at 64 sones. If you want a single number that directly reflects “how loud does this feel,” sones are the most intuitive unit.
Why Your Ears Hear Frequencies Differently
The equal loudness contours (originally mapped by Fletcher and Munson, later refined by ISO standards) reveal something striking about human hearing. Your ears are most sensitive to frequencies between roughly 2,000 and 5,000 Hz, a range that overlaps with speech consonants and alarm signals. This heightened sensitivity comes from the natural resonance of your ear canal, which amplifies sounds in that frequency band.
At soft volumes, this effect is dramatic. The 0-phon contour (the quietest sounds you can detect) rises steeply at low frequencies, meaning a bass tone needs far more physical energy than a midrange tone for you to hear it at all. At very loud levels, the curves flatten out considerably. Your hearing response becomes more uniform across frequencies when sounds are intense. This is why music can sound “thin” at low volumes but fuller when you turn it up: the bass frequencies that were below your perception threshold start to register.
Frequency Weighting: dBA and dBC
Sound level meters don’t just measure raw decibels. They apply frequency weighting filters that shape the measurement to match specific purposes. The two most common are A-weighting and C-weighting.
A-weighted decibels (dBA) filter the measurement to approximate how the human ear perceives sound, de-emphasizing very low and very high frequencies. This is the standard for workplace noise regulations, environmental noise monitoring, and most everyday sound measurements. When you see a noise level listed for an appliance or a concert, it’s almost always in dBA.
C-weighted decibels (dBC) capture a much flatter frequency response, including low-frequency energy that A-weighting filters out. This matters for controlling bass-heavy noise. Some venues use both limits simultaneously. The Hollywood Bowl, for instance, enforces a 95 dBA limit along with a separate 108 dBC cap specifically to control bass frequencies that would otherwise escape the A-weighted measurement.
Sound Level Meters
Professional sound measurement relies on dedicated sound level meters classified under the IEC 61672-1 international standard. These come in two accuracy tiers.
Class 1 meters cover a broader frequency range (16 Hz to 16 kHz), operate across wider temperature extremes (minus 10°C to 50°C), and often exceed the minimum 60 dB dynamic range required by the standard. They’re the choice for environmental noise assessment, building acoustics, and reverberation time measurements where precision is critical.
Class 2 meters cover a narrower frequency range (20 Hz to 8 kHz) and a tighter temperature window (0°C to 40°C), but they’re fully accepted for occupational noise monitoring under OSHA and ISO guidelines. For most workplace compliance checks, a Class 2 meter provides sufficient accuracy at a lower cost.
Smartphone Apps as Alternatives
NIOSH tested several smartphone sound measurement apps against calibrated reference equipment and found that three apps produced A-weighted readings within plus or minus 2 dBA of the reference measurements. When external microphones were connected to the phones, accuracy improved to within plus or minus 1 dB. That’s not lab-grade precision, but it’s close enough to give you a reliable sense of whether your environment is dangerously loud. For a quick check at a concert or on a factory floor, a well-reviewed sound meter app on your phone is a reasonable tool.
Safe Exposure Thresholds
NIOSH sets a recommended exposure limit of 85 dBA averaged over an eight-hour workday. For every 3 dBA increase above that, the safe exposure time cuts in half. So at 88 dBA, you should limit exposure to four hours. At 91 dBA, two hours. At 100 dBA, the safe window shrinks to minutes. Any noise above 85 dBA, regardless of duration, poses a risk to your hearing over time.
LUFS: Loudness in Digital Audio
If you work with music or video, loudness is measured in LUFS (Loudness Units relative to Full Scale). This standard was developed to normalize audio across broadcasts and streaming, preventing the jarring volume jumps that used to happen between TV shows and commercials or between songs on a playlist.
Each major streaming platform targets a specific LUFS level and automatically adjusts tracks to match. Spotify defaults to minus 14 LUFS, with user-selectable options at minus 11 and minus 19. YouTube also targets minus 14 LUFS. Apple Music uses minus 16 LUFS, which is slightly quieter. If audio is mastered louder than the platform’s target, it gets turned down automatically. This ended the “loudness wars” in music production, where tracks were compressed and boosted to sound louder than competing songs, often at the expense of audio quality.
Computational Loudness Models
For engineering and product design applications, loudness can be calculated mathematically from a sound’s frequency spectrum rather than measured by a listener. The most widely used approach is the Zwicker loudness model, standardized as DIN 45631 and ISO 532B. It takes a one-third octave band spectrum of a stationary sound and calculates its loudness in sones, accounting for how different frequency bands contribute to perceived volume under either free-field (outdoor) or diffuse-field (indoor) listening conditions. A separate ANSI standard (S3.4) provides an alternative calculation method. These models are used in product noise labeling, automotive cabin design, and appliance testing where consistent, repeatable loudness values are needed without assembling a listening panel.