Sound Pressure Level
Sound, or noise, is the term given to variations in air pressure that are capable of being detected by the human ear. Small fluctuations in atmospheric pressure (sound pressure) constitute the physical property measured with a sound pressure level meter. Because the human ear can detect variations in atmospheric pressure over such a large range of magnitudes, sound pressure is expressed on a logarithmic scale in units called decibels (dB). Noise is defined as “unwanted” sound.
Technically, sound pressure level (SPL) is defined as:
SPL = 20 log (P/Pref) dB
where P is the sound pressure fluctuation (above or below atmospheric pressure) and Pref is the reference pressure, 20 μPa, which is approximately the lowest sound pressure that can be detected by the human ear.
The sound pressure level that results from a combination of noise sources is not the arithmetic sum of the individual sound sources, but rather the logarithmic sum. For example, two sound levels of 50 dB produce a combined sound level of 53 dB, not 100 dB. Two sound levels of 40 and 50 dB produce a combined level of 50.4 dB.
A-Weighted Sound Level
Studies have shown conclusively that at equal sound pressure levels, people are generally more sensitive to certain higher frequency sounds (such as made by speech, horns, and whistles) than most lower frequency sounds (such as made by motors and engines) at the same level. To address this preferential response to frequency, the A-weighted scale was developed. The A-weighted scale adjusts the sound level in each frequency band in much the same manner that the human auditory system does. Thus the A-weighted sound level (read as “dBA”) becomes a single number that defines the level of a sound and has some correlation with the sensitivity of the human ear to that sound. Different sounds with the same A-weighted sound level are perceived as being equally loud. The A-weighted noise level is commonly used today in environmental noise analysis and in noise regulations. Typical values of the A-weighted sound level of various noise sources are shown in Table 1.
Table 1
Common Sound Levels in dBA
| Common Outdoor Sounds | Sound Pressure Level (dBA) | Common Indoor Sounds | Subjective Evaluation |
| Auto horn at 10’Jackhammer at 50’ | 100 | Newspaper pressTextile mill | Deafening |
| Gas lawn mower at 4’Pneumatic drill at 50’ | 90 | Auditorium during applauseFood blender at 3’ | Very Loud |
| Concrete mixer at 50’Jet flyover at 5000’ | 80 | Telephone ringing at 8’Vacuum cleaner at 5’ | |
| Large dog barking at 50’Large transformer at 50’ | 70 | Electric shaver at 1’Clothes washer at 2′ | Loud |
| Automobile at 55 mph at 150’Urban residential | 60 | Normal conversation at 3’Window air conditioning unit | |
| Birds at 25’Small town residence | 50 | Office noiseConference room background | Moderate |
| Wind in trees (5 mph)Farm valley | 40 | Soft stereo music in residenceLibrary | |
| Rustling leaves | 30 | Average bedroom at nightSoft whisper at 3’ | Faint |
| Quiet rural nighttime | 20 | Broadcast and recording studio | |
| 10 | Human breathing | Very Faint | |
| 0 | Threshold of hearing (audibility) | ||
Sensitivity to Changes in Sound Level
Human sensitivity to changes in sound pressure level is highly individualized. Sensitivity to sound depends on frequency content, background noise, time of occurrence, duration, and psychological factors such as emotions and expectations. However, in general, an approximation of human sensitivity to changes in sound level can be expressed as shown in Table 2. Noise is measured in decibels (dBA). Because people respond differently to sound at different frequencies, a weighted scale (dBA) is used to approximate the sensitivity of the human ear. Note that a 6 dBA change is required for the sound level change to be clearly noticeable.
Table 2
Human Sensitivity to Changes in Sound Level
| Change in Sound Levels (dBA) | Change in Apparent Loudness |
| 1 | Imperceptible |
| 3 | Just barely perceptible |
| 6 | Clearly noticeable |
| 10 | About twice (or half) as loud |
| 20 | About four times (or 1/4 as loud) |
Equivalent Sound Level
The Equivalent Sound Level (Leq) is a type of average which represents the steady level that, integrated over a time period, would produce the same energy as the actual signal. The actual instantaneous noise levels typically fluctuate above and below the measured Leq during the measurement period. The A-weighted Leq is a common index for measuring environmental noise.
Day-Night Average Sound Level
The day-night average sound level (DNL) descriptor is a 24-hour descriptor computed by averaging (on an energy basis) the hourly equivalent sound level (Leq) measured in each hour during a 24-hour period after 10 dB is added to the levels measured between 10 PM and 7 AM.
Frequency
Frequency (measured in Hz) is a measure of the cycles per second of a sound. With a sound level meter (or app on a smart phone) you can measure the sound level (in dBA). You can also look at the spectrum by taking that microphone signal and performing an FFT (Fast Fourier Transform). The spectrum will show you the amplitude versus frequency. This is helpful to know to determine how best to mitigate a noise source and for identifying noise sources.
There are other terms that I will describe as we go – as they pertain to certain projects such as:
- Noise exposure (for OSHA noise limits)
- Sound Transmission Class or STC
- Impact Insulation Class or IIC
- NC and RC that are used for mechanical equipment noise
- Statistical noise levels
- Reverberation time
Thank you for your interest in the Noise Engineers podcast.
Noise Engineers provides information and resources to help people address acoustical issues. In these episodes my goal is to provide resources, inexpensive tools, rules of thumb when dealing with acoustical issues. I would like to explain basic acoustic principles and answer any questions. I will describe actual projects to make this as practical as possible.
You can find our other podcasts at Noise Engineers podcast and iTunes
I welcome suggestions, comments, and questions. You can contact me on Facebook, Twitter , LinkedIn, email me ([email protected]) or call 520-979-2213.
