1.1 General This part of ISO 3743 specifies a relatively simple engineering method for determining the sound power levels of small, movable noise sources. The measurements are carried out when the source is installed in a specially designed room having a specified reverberation time over the frequency range of interest. The A-weighted sound power level of the source under test is determined from a single A-weighted sound pressure level measurement at each microphone position, rather than from a summation of octave-band levels. This direct method eliminates the need for a reference sound source, but requires the use of a special reverberation test room. The direct method is based on the premise that the sound pressure level, averaged in space and time in the test room, can be used to determine the sound power level emitted by the source. The properties of the special reverberation test room are chosen so that the room\'s influence on the sound power output of the equipment under test is small. The number of microphone positions and source locations required in the test room are specified. Guidelines for the design of special reverberation rooms are given in annex B. In addition to the direct method, a comparison method is also described (see 8.3). However, since the requirements on the test room for the comparison method of ISO 3743-1 are considerably less restrictive, it is recommended that the comparison method of ISO 3743-1 be used if a special reverberation test room is not available. NOTE 1 Precision methods for the determination of the sound power levels of small noise sources are specified in ISO 3741 and ISO 3745. 1.2 Types of noise The methods specified in this part of ISO 3743 are suitable for measurements of all types of noise within a specified frequency range, except impulsive noise consisting of isolated bursts of sound energy. NOTES 2 A classification of different types of noise is given in ISO 12001. 3 For sources of impulsive noise consisting of short-duration noise bursts, the free-field methods specified in ISO 3744 and ISO 3745 should be used. 1.3 Noise source The noise source may be a device, machine, component or sub-assembly. The maximum size of the source under test and the lower limit of the frequency range for which the methods are applicable depend upon the size of the room used for the acoustical measurements. The volume of the noise sources should not exceed 1 % of the volume of the special reverberation test room. For the minimum test room volume of 70 m3, the recommended maximum size of the source is 0,7 m3. Measurements on sources emitting discretefrequency components below 200 Hz are frequently difficult to make in such small rooms. 1.4 Measurement uncertainty Determinations made in accordance with this part of ISO 3743 result, with few exceptions, in standard deviations of reproducibility equal to or less than 2,0 dB from 500 Hz to 4 000 Hz, 3,0 dB for 250 Hz and 8 000 Hz, and 5,0 dB for 125 Hz (see table 1). A single value of the sound power level of a noise source determined according to the procedures of this part of ISO 3743 is likely to differ from the true value by an amount within the range of the measurement uncertainty. The uncertainty in determinations of the sound power level arises from several factors which affect the results, some associated with environmental conditions in the measurement laboratory and others with experimental techniques. If a particular noise source were to be transported to each of a number of different laboratories, and if, at each laboratory, the sound power level of that source were to be determined in accordance with this part of ISO 3743, the results would show a scatter. The standard deviation of the measured levels could be calculated (see examples in ISO 7574-4:1985, annex B) and would vary with frequency. With few exceptions, these standard deviations would not exceed those listed in table 1. The values given in table 1 are standard deviations of reproducibility, σR, as defined in ISO 7574-1. The values of table 1 take into account the cumulative effects of measurement uncertainty in applying the procedures of this part of ISO 3743, but exclude variations in the sound power output caused by changes in operating conditions (e.g. rotational speed, line voltage) or mounting conditions. The measurement uncertainty depends on the stan dard deviation of reproducibility tabulated in table 1 and on the degree of confidence that is desired. As examples, for a normal distribution of sound power levels, there is a 90 % confidence that the true value of the sound power level of a source lies within the range ± 1,645σR of the measured value and a 95 % confidence that it lies within the range ± 1,96σR of the measured value. For further examples, reference may be made to the ISO 9296 and ISO 7574 series. NOTES 4 The standard deviations listed in table 1 are associated with the test conditions and procedures defined in this part of ISO 3743, and not with the noise source itself. They arise partly from variations between measurement laboratories in the geometry of the test room, the acoustical properties of the test room boundaries, background noise, the type and calibration of instrumentation, and the reference sound source. They are also due to variations in experimental measurement techniques, including microphone placement and spatial averaging, location of source under test, integration times, and measurement of reverberation time. 5 If several laboratories use similar facilities and instrumentation, the results of sound power determinations on a given source in those laboratories may be in better agreement than would be implied by the standard deviations given in table 1. 6 For a particular family of sound sources, of similar size with similar sound power spectra and similar operating conditions, the standard deviations of reproducibility may be smaller than the values given in table 1. Hence, a noise test code for a particular type of machinery or equipment making reference to this part of ISO 3743 may state standard deviations smaller than those listed in table 1 if substantiation is available from the results of suitable interlaboratory tests. 7 The standard deviations of reproducibility, as tabulated in table 1, include the uncertainty associated with repeated measurements on the same noise source under the same conditions (for standard deviation of repeatability, see ISO 7574-1). This uncertainty is usually much smaller than the uncertainty associated with interlaboratory variability. However, if it is difficult to maintain stable operating or mounting conditions for a particular source, the standard deviation of repeatability may not be small compared with the values given in table 1. In such cases, the fact that it was difficult to obtain repeatable sound power level data on the source should be recorded and stated in the test report. 8 The procedures of this part of ISO 3743 and the standard deviations given in table 1 are applicable to measurements on an individual machine. Characterization of the sound power levels of batches of machines of the same family or type involves the use of random sampling techniques in which confidence intervals are specified, and the results are expressed in terms of statistical upper limits. In applying these techniques, the total standard deviation must be known or estimated, including the standard deviation of production, as defined in ISO 7574-1, which is a measure of the variation in sound power output between individual machines within the batch. Statistical methods for the characterization of batches of machines are described in ISO 7574-4. Table 1 - Estimated values of the standard deviation of reproducibility of sound power levels determined according to this part of ISO 3743 Octave-band centre frequency Hz Standard deviation of reproducibility, σR dB 125 250 500 to 4 000 8 000 5,0 3,0 2,0 3,0 A-weighted 2,01) 1) Applicable to a source which emits noise with a relatively \'flat\' spectrum in the frequency range 100 Hz to 10 000 Hz.