Whether indoors or out, in good conditions or bad, some problems are common to almost all noise measurements.
The convenient measurable range of sound levels extends from about 10 dB above the ambient level (which can be as low as 20 dB in good test rooms) to about 140 dB or so. Whenever measurements are made with background noise within 10 dB of the test noise level, errors tend to increase. By whatever means is convenient, the difference between the test noise level and background should be increased. Work outside these limits is possible, but not easy.
The analyst can increase the sensitivity of the measuring instrument at low levels by using a preamplifier (but with danger of introducing electrical noise) or reduce it at high levels with attenuators (but the mechanical vibration in high-noise fields can affect the operation of the instrument).
Microphones vary in their characteristics. Condenser types have wide frequency response, good sensitivity, and withstand high temperatures and high-sound levels. They are often deranged by high humidity, they require some auxiliary equipment; and their cost is rather high.
Piezoelectric microphones (made with synthetic ceramic) are excellent for general purpose use. Their stability and sensitivity are good, but their frequency response is less wide than that of condenser microphones. Their cost is relatively low. They are not usable at extremely low or extremely high temperatures.
Dynamic microphones can be used at both lower and higher temperatures than the piezoelectric and at higher humidities than the condenser types. They sometimes lose calibration over long periods of time, however, and their frequency response can vary from temperature change. Their construction employs acoustic resonating chambers, which must remain clean and dust free. None of the three types is completely nondirectional.
Frequently the acoustic power of a source is needed but is not supplied by the maker. Ideally this kind of data should be determined in a quiet room or in a free-field situation with low background noise, but practical values can often be approximated by field measurements. For outside sources, if reflecting surfaces (except the ground or pavement) are not close by and the device is the principal noise source, the following equation gives the power level:
[See also Section 6.1 and Equation 6.1(9).] Here r is the distance in feet from the source to the sound pressure meter, and Q is a directivity factor.
The directivity factor varies from about 2 for a hard pavement to about 0.5 for absorbing grass; the sound radiated by the machine is usually somewhat directional in itself, too. Several SPL measurements should be taken to strike an average. Values obtained in this way are seldom highly accurate; usually field measurements give low values for PWL. A simpler procedure (for rough estimates) is to measure the SPL at a distance of 1 meter; the PWL value is about 15 units higher. (This procedure can be used on small sources only.) Neither of these procedures works well on extended sources such as cooling towers, large trucks, or the like, nor for distances greater than about 60 meters.
The term here called directivity factor is sometimes called directional gain or DG. It is the factor by which the power of the source should be multiplied, if its sound radiation were nondirectional, to give the measured level in the direction of measurement. The directivity factor must be determined for the location and direction in question for each measurement or machine.
Field measurements can be disturbed by wind noise. Analysts can reduce this problem by using a windscreen. Windscreens have various forms, but a common method is to surround the microphone with a skeletal spherical frame about 15 to 20 cm in diameter and stretch fine-meshed cloth over it.
At extreme distances, sound behavior becomes unpredictable; but noise control work seldom requires measurements beyond a mile or so. At high altitudes, low barometric pressure, or high humidity, the density change in the atmosphere affects calculations. For most work, these effects can be ignored.
Multiple sources are usually present, as well as multiple transmission paths; and as in other measurements, the presence of the measuring device can affect the local conditions. Interpretation in both raw and tabulated data can be a problem. Judgment, based on experience, is of the greatest importance in noise-control work.
Was this article helpful?