HOME | Site Map | Legal Notes | Acknowledgements | Contact Us

Sustainable Aggregates Logo

 

information gateway

Noise

 

Units of Measurements


There are a number of excellent web sites which give good descriptions of the key parameters in noise and noise monitoring. A practical guide is given in www.noisenet.org while a more academic but highly interactive treatment of sound is given in Acoustics and Vibration Animations. The recent version of the British Standard giving a code of practice for the control of noise from open sites relating to control of noise (BS 5228-1:2009 445) also contains useful technical and background information

The descriptions which follow are designed to provide the non-expert with a basic understanding of the terms and measures involved.

Pascals (Pa) and Decibels (dB)

Sound has two characteristics; amplitude and frequency. The amplitude is the amount of pressure exerted by the air and is usually measured in Pascals (Pa) or microPascals. Table 1 shows that the values cover a vast range, and so they are usually converted into decibels (dB) which is a logarithmic scale and takes 20 microPascals (threshold of hearing) as its reference level. Table 1 gives Pa and equivalent dB values for a number of different activities.

Table 1. Sound pressure and level, together with examples

Sound pressure (microPascals)

Sound level (dB)

Example

200,000,000

140

threshold of pain

 

130

riveting on steel plate

20,000,000

120

pneumatic drill

 

110

loud car horn at 1m

2,000,000

100

alarm clock at 1m

 

90

inside underground train

200,000

80

inside bus

 

70

street-corner traffic

20,000

60

conversational speech

 

50

business office

2000

40

living room

 

30

bedroom at night

200

20

broadcasting studio

 

10

normal breathing

20

0

threshold of hearing (@ 1kHz)

 

Frequency (Hz)

Frequency refers to how quickly the air vibrates and is measured in Hertz (Hz) as a sinusoidal wave. It is subjectively felt as the pitch of the sound. Most of the time, noise is made up of different amounts of a broad range of frequencies, which are constantly changing. Generally, the lowest frequency audible to humans is 20Hz and the highest is around 20,000Hz, although this usually decreases with age.

Noise (dB(A))

When measuring environmental noise (the sounds heard by the person concerned), the equipment can use a weighting function which filters the frequency of sound to mimic the characteristics of human hearing. This produces readings expressed as dB(A) and the scale remains logarithmic. Two machines emitting exactly the same noise level of 80dB(A) produce a total noise of 83dB(A), not 160dB(A). A 10dB(A) increase in sound level represents a doubling of loudness. An average living room would typically have a noise level of about 40dB(A) while busy road traffic would generate about 70-80dB(A) measured on the pavement (approximately 3m from the vehicles).

Between 1kHz and about 7kHz, the weighting makes very little difference, so the level in dB(A) is very similar to the level in dB (often reffered to as dB(Lin)). Outside this range the difference between the two measurements increases rapidly.

Measuring noise (SPL)

Noise emissions are measured using sound-level meters (SLM), which detect and record changes in the sound pressure level (SPL). As the level is constantly changing, integrating SLMs are used to obtain various "averages" and other descriptors of interest (e.g. LAeq,T, LAmax,T, LAmin,T, LA90,T which are described and displayed in Figure 1). These values can be used to describe the data which may have been recorded over several minutes or hours, and so is likely to be varied and complex. The descriptors always have the time interval defined (T in the examples above), over which that description applies. With environmental monitoring it is usually 1 hour.

Flash - Interactive diagram showing different noise measurements

Figure 1. Interactive diagram showing different noise measurements.

Select: Flash version or HTML5 version.

 

Noise from a particular source will be reflected by any facade that directly faces that source. Thus a microphone 1-2m in front of a building would typically yield a level 3dB(A) higher than a free-field measurement (i.e. at least 3.5m away from a facade) because the reflected wave would be added to the direct wave.

Background noise levels should be established by continuous monitoring over a period sufficient to provide a representative picture of the noise environment or by averaging results from short sampling periods. Once the noise source has started, BS 5228-1:2009 445 suggests the background noise level (more correctly termed the residual noise level) can be considered as the sound pressure level that is exceeded for 90% of the given time interval (LA90, T), although this does depend on the type of noise.

Noise sources (SWL)

To understand the noise environment around a site, each source of noise must be identified and measured. The level of sound generated by each source can be defined as its Sound Power Level (SPL). This should be provided by the manufacturer, but Annexes C and D of BS 5228-1:2009 445 provide generalised data (current and historic) on noise emissions from various plant and activities. If the SWL cannot be obtained, then it can be determined by measuring the SPL at a distance of 10m from the source and using the following equation. It is possible that the noise level might vary in different directions from a piece of equipment, in which case an average could be taken, or if it is a particularly sensitive issue, different SWLs could be used for predicting noise levels in different directions from the source

SWL = SPL + (20*Log10(Distance)) + 8
  = SPL + (20*Log10(10)) + 8
  = SPL + 20 + 8

 

Noise Prediction

Noise prediction requires the combination of noise from each item of plant/activity to arrive at the equivalent continuous sound level (LAeq,T ) after taking account of the noise generated, the amount of time in use, the distance of the reception point from the noise-generator and whether there will be any screening.

If a single noise source is considered, the predicted noise level can be calculated at specific location if the SWL of the source, and the distance between source and receptor are both known.

SPL = SWL - (20*Log10(Distance)) - 8

However, the measured SPL is likely to be lower than this, as adjustments have to be made for the presence of any barriers, whether there is any reflection involved and the nature of the ground the noise has to travel over. Other factors such as meteorological conditions (particularly wind speed and direction) and atmospheric absorption can also influence the level of noise received, and add greatly to the complexity.

The corrections for barriers require knowledge of sound pressure levels at different frequencies and of the precise geometry of the receptor in relations to the source and barrier. These adjustments are quite complex, and are outlined in detail in Annex F of BS 5228-1:2009 445. However the Standard suggests that, as a working approximation, an attenuation of 10dB can be made if a screen completely hides the source from the receiver. If the top of the plant is just visible to the receiver, then an attenuation of 5dB can be assumed.

The adjustments for ground conditions are also complex and require the classification of the ground as either hard or soft. Further details should be obtained from the British Standard 445.

Various techniques exist 447 448 that attempt to accurately predict noise levels by drawing all the factors together. However at distances over 300m (the usual case for quarries) noise predictions should be treated with caution, especially where a soft ground correction factor has been applied, because of the increasing importance of meteorological effects.

Multiple sources

In most instances, a receptor will be subject to noise from a number of different sources. If this is the case, then the total SPL at that location can be obtained from the SPLs of each source by the following equation.

SPL = 10*log10[10 SPL1/10 + 10 SPL2/10...+ 10 SPLn/10]

An example was given earlier where the result of adding two sources of noise of 80dB(A) gave a resulting SPL of 83dB(A). Where two or more sources with very different noise levels are combined, the combined level will be just slightly higher than the highest value. For example, two sources giving 68dB(A) and 56dB(A) at a location would give a combined value of:

SPL = 10*Log10(10(68/10) + 10(56/10))
  = 68.3 dB(A)

 

Moving Sources

For mobile equipment the factors used are slightly different. These can sometimes be treated as line sources, so not only has the SWL of the equipment and the distance to be taken into account, but also the speed of the machine or the flow rate. The equation used is then:

SPL = SWL - 33 + 10*log10(Flow rate) - 10*log10(Velocity) - 10*log10(Distance)

Further corrections are outlined in Annex F of BS 5228-1:2009 445 and include factors for angle of view, reflections from facades, etc.

Noise Modelling

When there are multiple static or moving sources of noise on a site (as is usually the case) then noise modelling software can be used to produce noise contour maps around the site, or to calculate noise at specific receiver locations. This can be particularly useful at the quarry design stage, when the potential noise impact is being assessed, and when changes to the quarry design can still be incorporated, should problems be highlighted. They cannot take account of changing meteorological and atmospheric conditions, so there will always be a margin of error in their predictions. Examples of this type of modelling software include Sitenoise (part of the NoiseMap environmental noise mapping software - www.noisemap.ltd.uk) and SoundPLAN (www.soundplan.eu).

 

Continued with Potential Sources of Noise TOP Return to Introduction