Fletcher-Munson Curves Explained: Equal-Loudness Contours, Phons, and Why Volume Changes Tone

Fletcher-Munson curves are a set of equal-loudness contours that map how much sound pressure level (SPL) each frequency needs to sound equally loud to the average human listener. The ear is brutally uneven: at conversational listening levels, a 100 Hz bass tone must be roughly 5 dB louder in physical SPL than a 1 kHz tone to sound the same volume, a 50 Hz tone about 11 dB louder, and a 20 Hz tone nearly 30 dB louder. This is why music played at low volume sounds thin and clappy, and why it seems to “fill out” when you turn it up: your ear’s bass sensitivity improves as level rises.

The practical consequence is that perceived tone is a function of level, not just frequency content. Before trusting any EQ decision, know the SPL your monitors are actually producing at the listening seat with the free browser-based Online Decibel Meter. It plots live level readings so you can stop mixing by guesswork.

---

What the Curves Actually Say

An equal-loudness contour is a curve of SPL versus frequency that yields constant perceived loudness. Each curve is labeled in phons, and the definition is precise: the phon value of any tone equals the SPL in dB of an equally loud 1 kHz pure tone. A 100 Hz tone that sounds as loud as a 40 dB SPL 1 kHz tone is therefore assigned 40 phons, even though its actual SPL is around 45 dB.

The numbers below are rounded approximations of the current ISO 226 standard curves (frontal loudspeaker listening, young normal-hearing subjects):

Loudness level (1 kHz reference)	50 Hz needs	100 Hz needs	4 kHz needs
80 phon	≈ +4 dB (84 dB SPL)	≈ −1 dB (79 dB SPL)	≈ −5 dB (75 dB SPL)
60 phon	≈ +8 dB (68 dB SPL)	≈ +3 dB (63 dB SPL)	≈ −5 dB (55 dB SPL)
40 phon	≈ +11 dB (51 dB SPL)	≈ +5 dB (45 dB SPL)	≈ −6 dB (34 dB SPL)

Two structural facts matter most in practice:

1. The contours flatten as level rises. At 80 phons, the bass deficit across 50 Hz to 10 kHz shrinks to a few dB. At 40 phons (a quiet living room), the same span swings by nearly 40 dB, including a ~30 dB hole at 20 Hz.
2. The ear is most sensitive around 2 kHz to 5 kHz. The ear canal is roughly a 2.5 cm to 3 cm closed tube; its quarter-wave resonance ($f \approx c / 4L \approx 343 / (4 \times 0.027) \approx 3.2$ kHz) plus middle-ear ossicle transfer gives this region up to ~6 dB of free gain over 1 kHz at every loudness level.

Phon Is Not Sone

The phon is a loudness level read off the contours; the sone is a linear loudness scale. By convention, 1 sone is defined as 40 phons, and perceived loudness doubles with every +10 phon increase above that point: 50 phons sounds twice as loud as 40 phons, 60 phons four times as loud. That 10 dB-per-doubling figure only holds on average and mainly in the mid-range; the perceptual mapping also depends on bandwidth and duration, which is why the pure-tone contours are only an approximation for real music.

Related energy math worth keeping straight: +3 dB doubles acoustic power, +10 dB doubles perceived loudness (for broadband signals). A mixing fader move of 3 dB is an engineering event; a move of 10 dB is what a listener actually calls “twice as loud.”

A Short Standards History (and What to Cite in 2026)

Harvey Fletcher and W.A. Munson of Bell Laboratories published the original curves in 1933 in the Journal of the Acoustical Society of America (“Loudness, its definition, measurement and calculation,” JASA 5, 82–108), based on headphone listening trials. Robinson and Dadson re-determined the data in 1956 using loudspeakers, and that work became the basis of the first ISO 226 standard. ISO revised the curves in 2003 after an international pooled study (heavily contributed by Tohoku University), and again in 2023. As of 2026, ISO 226:2023 (Edition 3, published March 2023) is the current revision — if you see a 20 Hz “loudness compensation” chart copied from a 1933 paper, it is a historical artifact, not the modern reference. The A-weighting curve used by most sound level meters was historically anchored to the 40-phon contour, which is why dBA readings disagree with subjective loudness for strong bass content.

Why Engineers Calibrate Near 85 dB SPL

Reference monitoring at ~85 dB SPL per channel (measured C-weighted, slow, at the listening seat) sits in the flattest region of the equal-loudness family. Between roughly 80 and 90 phons, the bass-to-midrange disparity drops to only a few dB, so decisions made on balance, bass level, and vocal presence transfer reasonably to both loud club systems and quiet car stereos. Below about 70 dB SPL, you will systematically over-EQ the low end to compensate for your own ear’s roll-off, then hear the result as flabby at normal volume.

Two honest caveats:

• 85 dB SPL continuous is at the boundary of safe long-duration exposure. NIOSH’s recommended exposure limit is 85 dBA for 8 hours with a 3 dB exchange rate (88 dBA halves allowable time to 4 hours). OSHA’s action level is 85 dBA and its permissible limit 90 dBA. Commercial cinema alignments, by contrast, span roughly 75 to 85 dB SPL per channel depending on format and room size, so near-field monitoring at 85 dB SPL sits at the loud end of pro practice.
• Near-field studios are often calibrated to 79 to 83 dB SPL precisely because small rooms have strong modal behavior (see the companion acoustics guides), and 85 dB in a untreated 9 m² room excites every mode hard.

Practical Workflow With a Browser Meter

1. Establish your seat level. Play pink noise at −18 dBFS from your DAW, hold the phone or laptop microphone at ear height, and read the Decibel Meter while adjusting monitor gain until the meter’s typical reading matches your target zone (80 to 85 on an uncalibrated consumer microphone roughly corresponds to calibrated ~85 dB SPL on many built-in MEMS microphones, but this is device-dependent).
2. Check the low-volume story. Re-run a familiar reference track at a conversational level (~60 dB SPL) and notice the bass and air recede first. That is the 60-phon contour, not your mix being “warm.”
3. Hear the contours directly. Our Perceptual Sine Sweep sweeps level-compensated tones so you can perceive where your own hearing deviates from the standard curves.
4. Account for your ears, not just the standard. ISO 226 describes 18-to-25-year-old normal hearing. High-frequency sensitivity declines predictably with age; run a browser hearing test and read Normal Human Hearing Range before blaming the room, the monitors, or the curve.

Limits, Myths, and Troubleshooting

• Contours are pure-tone data. Music with sustained bass content, broad bandwidth, and critical-band integration does not follow the curves exactly; treat them as a ±5 dB guide, not a law.
• A browser microphone is not an SPL meter. Web Audio exposes raw digital levels (dBFS-relative), and phone microphones apply unknown device-specific sensitivity and smoothing. The Online Decibel Meter gives consistent relative readings; only a calibrated class 2 sound level meter gives legal-grade dB SPL or dB(A).
• Headphones do not follow loudspeaker contours. Equal-loudness data for headphone presentation differ because there is no head shadow or pinna filtering; loudness-compensated “Loudness Equalizer” presets tuned on speakers will misbehave on IEMs.
• Small speakers physically cannot exploit the low end of the curves. A phone or laptop that cannot radiate below ~200 Hz cannot demonstrate the 20 Hz cliff at all; see Why Smartphones Cannot Reproduce 20 Hz.
• The curves do not say “turn everything up.” A cheap workaround for thin low-volume listening is a loudness-contour EQ (a 6 to 10 dB shelf at 100 Hz) on the playback device, not permanently raising SPL, which trades tone for noise dose.

The equal-loudness contours are the most useful psychoacoustic chart for anyone who mixes, masters, or tunes a home theater: they explain, within a few decibels, why tone judgments change when the volume does.