Like other animals, humans hear different pitched sound differently. Some sounds, such as those in the range of speech or a baby crying, are easy to hear and emphasized by the ear. Other sounds, such as low-pitched rumbling or high-pitched buzzing might be more difficult to hear. In an effort to take these differences into account, acoustics professionals have developed patterns for emphasizing or de-emphasizing sounds at different pitches. In this article, we will discuss these patterns—called frequency weighting curves—and how they are used in acoustics.
In more technical terms, the pitch of a sound is its frequency (how many cycles the sound wave does each second.) The frequency weighting curves add or subtract from the sound level on a frequency-dependent basis. How much a frequency is weighted depends on the type of curve applied. All frequency weighting curves are named with letters of the alphabet.
A-weighting is by far the most widely used frequency weighting used in acoustics. It de-emphasizes low and high frequencies while slightly boosting middle frequencies (around 1–4 kHz.) A-weighting is based on the historical equal-loudness contours. This means that A-weighting is meant to represent what humans are capable of hearing. When the A-weighting curve was created, it was meant for quiet sounds with sound pressure levels (SPL) of about 40 dB. However, with time A-weighting has been applied to many situations, even when such applications might not be an ideal fit for A-weighting, at least from a strictly scientific point of view.
There are several advantages to A-weighting. A weighting is the standard used for workplace safety regulations both nationally and internationally. This makes it great for comparing data from all over the world and for comparing old data to new data. Additionally, most sound level meters use A-weighting as a standard setting, meaning you don’t have to do extra processing of data to get A-weighted levels, which might not be true of some of the other, less common frequency weighting curves.
B-weighting, which falls between A- and C-weighting, is no longer in use, has little practical use. We’re not going to spend time on it here.
C-weighting is the next most common type of frequency weighting used following A-weighting. C-weighting is generally flat, dropping off at the highest and lowest frequencies. C-weighting is less severe on low frequencies than A-weighting and represents the response of the human ear to loud sounds (over 100 dB). C-weighting is more typical of machine noise and jet noise measurements than in other acoustics applications. All Class 1 sound level meters are mandated to allow for C-weighting.
D-weighting, like B-weighting, is no longer in use. D-weighting was meant for jet noise.
Z-weighting stands for Zero-weighting. It’s a bit of a misnomer to call Z-weighting a type of frequency weighting, since Z-weighting means “no weighting.” It is the actual level of the sound as measured by the microphone. Like all transducers, microphones have an ideal dynamic range. Systems that use Z-weighting are designed with a flat response over the frequencies of interest (usually from 20–20,000 Hz.)
When a sound pressure level is reported with frequency weighting, the letter of the weighting is attached to the end. For example, A-weighted decibels are written dBA or dB(A). The other letters follow the same pattern.
The letter of the weighting can also be found as the second letter in level abbreviations. For example, LAS means A frequency weighted sound levels with slow time weighting (read more about time weighting here.)
AcousticalEngineer.com has several tools to help with frequency weighting. Check them out here: