Frequency
The pressure variations associated with many sounds have a repeating pattern. This is especially true of musical sounds. The frequency of a repetitive or periodic phenomenon is the number of times per second that the pattern repeats itself. High frequency or high-pitched sounds are heard when the pressure variations in the air occur rapidly--several thousand times per second. The unit used for frequency is hertz (Hz). Thus a musical note with a frequency of 1200 Hz has a fundamental pattern of pressure variations that repeats itself 1200 times per second. In building acoustics, it is important to know the frequencies that make up a sound because different frequencies behave differently.
Typical sounds in buildings are complex without any clearly defined pattern, and are usually classified as noise. There is usually no obvious frequency or tone associated with such noises, although there may be tonal characteristics such as "rumbling" or "hissing". Despite this, it is possible to analyse noises in terms of the energy contained in certain frequency intervals. This is called spectral analysis.
A narrow band spectrum shows the energy present at each frequency in great detail. This kind of information is very useful in some instances but more often the detail is overwhelming; in building acoustics it is common to present spectra in standard one-third-octave band or even octave band plots. The term octave has the same meaning in building acoustics as it does in music; adjacent octave bands have frequencies that differ by a factor of two. Adjacent one-third-octave bands are one-third of an octave apart. To produce this kind of spectrum the energy in specified regions of the narrow band spectrum is added together to give the octave or third-octave level. This type of presentation gives information that is easier to relate to the response of the ear.
The one-third-octave band spectrum for a particular noise source provides a good deal of information but a single number rating is often more convenient. One way of obtaining a single number describing a complex noise is to use A-weighted levels. The human ear is not equally sensitive at all frequencies; sounds of the same level but with different frequencies will not be considered equally loud. A sound at 3 kHz at a level of 54 dB, for example, will sound as loud as one at 50 Hz at a level of 79 dB.
The output of a microphone or sound level meter can be altered using an A-weighting electrical filter so that it more closely represents the response of the human ear. The resulting sound pressure levels are expressed as a number followed by the symbol dBA.
Noises with identical A-weighted levels can have quite different spectra and can evoke quite different responses from people. The use of such pejorative terms as whine, rasp, grate, rumble and hiss to describe sounds shows that people are well aware of the differences in spectral content. Other terms in common use such as loudness, noisiness and annoyance are not at all synonymous, nor clearly related to sound pressure levels. Despite the shortcomings of this simplistic method of rating noises, it is in common use.
This article continues with Acoustic Terms.