A decibel (dB) is a logarithmic unit used to express the ratio between two values of a physical quantity — in acoustics, most commonly sound pressure or sound power. It is not a fixed unit like metres or kilograms; it is a ratio scale, always measured relative to a reference value. When we say a sound is "60 dB," we mean its sound pressure level is 60 decibels above the reference threshold of human hearing.
The decibel scale compresses the enormous range of sound pressures the human ear can detect — from the faintest whisper to a jet engine, a factor of 10 million — into a manageable range of roughly 0 to 140. Without logarithms, acoustic engineering would drown in unwieldy numbers.
Real-World Analogy
Think of the Richter scale for earthquakes. A magnitude 5 earthquake is not twice as powerful as a magnitude 2.5 — it is about 5,600 times more energetic. The Richter scale uses logarithms to compress an enormous range of earthquake energies into single-digit numbers that humans can compare intuitively.
The decibel scale does the same thing for sound. A sound at 80 dB is not "twice as loud" as one at 40 dB — it has 100 times the sound pressure and 10,000 times the sound energy. The logarithmic scale turns these huge ratios into simple numbers you can add and subtract.
Technical Definition
The decibel is defined as one-tenth of a bel (named after Alexander Graham Bell). For sound pressure level (SPL):
L_p = 20 x log10(p / p_ref)
Where p is the measured sound pressure in Pascals and p_ref = 20 muPa (20 x 10^-6 Pa), the internationally standardised reference corresponding to the approximate threshold of hearing at 1000 Hz, per IEC 61672-1:2013.
For sound power level:
L_w = 10 x log10(W / W_ref)
Where W_ref = 10^-12 watts (1 picowatt).
The factor of 20 in the pressure formula (versus 10 in the power formula) arises because sound power is proportional to the square of sound pressure, and log(x^2) = 2 x log(x).
Key dB Relationships
These relationships are worth memorising:
| Change | Meaning |
|---|---|
| +3 dB | Sound energy doubles |
| +6 dB | Sound pressure doubles |
| +10 dB | Sound is perceived as roughly twice as loud |
| +20 dB | Sound pressure increases by a factor of 10 |
| +40 dB | Sound pressure increases by a factor of 100 |
Adding Decibels
Because decibels are logarithmic, they do not add arithmetically. Two identical sources at 60 dB each do not produce 120 dB — they produce approximately 63 dB. The formula for adding two incoherent sound levels is:
L_total = 10 x log10(10^(L1/10) + 10^(L2/10))
This means that adding a second identical source raises the level by only 3 dB. And if one source is 10 dB louder than another, adding the quieter source changes the total by less than 0.5 dB — it is essentially inaudible next to the dominant source.
Frequency Weighting
Raw dB SPL values treat all frequencies equally, but human hearing is not equally sensitive at all frequencies. A-weighting (dB(A)) applies a filter that mimics the ear's reduced sensitivity to low and very high frequencies, producing a single number that correlates better with perceived loudness. Most noise regulations and occupational health standards specify limits in dB(A).
C-weighting (dB(C)) applies a flatter filter and is used for peak sound levels and low-frequency noise assessment.
Why It Matters for Design
Decibels are the language of acoustic design. Every specification, every standard, every measurement, and every calculation result is expressed in decibels:
Reverberation time. RT60 is defined as the time for SPL to decay by 60 dB. The "60" in RT60 is a decibel value.
Noise criteria. NR, NC, and RC curves specify maximum dB SPL at each octave band. Compliance means your measured dB values are below the curve at every frequency.
Transmission loss. STC and Rw ratings describe how many dB a partition reduces sound pressure. A wall with STC 50 reduces transmitted sound by 50 dB (at the frequencies that matter most for speech).
Speech intelligibility. The signal-to-noise ratio (SNR), expressed in dB, is the primary determinant of STI. Every additional dB of SNR improves intelligibility until the SNR exceeds about 15 dB, above which further improvement yields diminishing returns.
Design targets. Background noise targets for rooms are specified in dB: 25-30 dB(A) for a recording studio, 30-35 dB(A) for a bedroom, 35-40 dB(A) for an office, 40-45 dB(A) for a restaurant.
A common mistake is treating dB as a linear unit. "Reducing noise by 50%" does not mean subtracting half the dB value. Halving the sound pressure reduces the level by 6 dB. Halving the perceived loudness requires about a 10 dB reduction — which means reducing sound energy by 90%.
How AcousPlan Uses This
Every output in AcousPlan is expressed in decibels. RT60 is the time for a 60 dB decay. Noise criteria compliance is evaluated against dB SPL thresholds at each octave band. Transmission loss values in the STC/Rw calculator are in dB. The speech privacy calculator expresses received speech levels, background noise, and signal-to-noise ratios in dB.
When you view the results dashboard, the frequency-dependent RT60 curve represents the decay behaviour at each octave band — and the underlying physics at every point on that curve is governed by how many dB of energy are absorbed at each reflection.
The compliance indicators show green when your dB values meet the standard and red when they exceed it — giving you an immediate, frequency-by-frequency picture of your room's performance in the universal language of acoustics.
Related Concepts
- What is Sound Pressure? — The physical quantity that dB SPL measures
- What is Noise? — Unwanted sound, quantified in dB
- What Are Octave Bands? — How dB values are broken down by frequency
- What is Transmission Loss? — Sound reduction measured in dB
- What is Frequency in Acoustics? — The property across which dB values vary
Calculate Now
Put decibels to work in your design. Use the AcousPlan Room Calculator to calculate RT60, check noise criteria compliance, and evaluate speech intelligibility — all expressed in the universal dB scale.