Machinery Noise Guide: Sound power measurement uncertainty research
Summary of NPL Research:
delta slm
This is the uncertainty resulting from errors associated with the use of instrumentation (Sound Level Meters).
Research Summary
Uncertainty components associated with use of a Sound Level Meter (SLM), have been identified and quantified in order to evaluate a suitable value for the slm uncertainty component for use with ISO 3744 uncertainty budget. Historical data from SLM parameter testing to BS 7580 was was analyzed. The study was restricted to 22 SLM types submitted to NPL for verification between 1997 and 2001.
Results
A summary of the the recommended A-weighted standard uncertainty value for delta slm, based on the analysis of the SLM data is as follows:
| To report a sound pressure measurement | 0.4 |
| To use in a sound power uncertainty budget | 0.33 |
Recommendations
A the recommendations from the research can be summarized as follows:
To reduce measurement uncertainty:
|
Further information
The report detailing this research is available here: 
NPL report DQL_AC2 (330 kB)
delta rep
This is the uncertainty resulting from temporal variability in the operating conditions of the noise source.
Research Summary
The aim of this research was to evaluate the uncertainty component, delta rep or repeatability uncertainty due to variability of the machine operating condition over time.
Short term variability with time, of the sound power output of a range of outdoor machine types was analyzed, based on the machine modes of operation described in the appropriate test codes
Results
A summary of the the recommended A-weighted standard uncertainty value for delta rep resulting from the measurements, is as follows:
| Fixed continuous operation mode |
0.2 |
| Activity mode |
1.0 |
Recommendations
The recommendation from the research can be summarised as:
| The above values for delta rep should be incorporated into ISO 3744 to apply to small machines |
Further information
The NPL report entitled 'Measurement uncertainties associated with sound power level determination due to source operating conditions' is available here: NPL Report DQL_AC6 (321 kB)
delta boun
This is the uncertainty resulting from errors due to sound reflections from boundaries within the test environment.
Research Summary
For ISO 3744, where environmental correction k2 is not applied due the assumption that the environment is essentially free field over a reflecting plane, the uncertainty component delta boun is assumed to be negligible. NPL has conducted research to evaluate delta boun in order to validate this assumption. Two research projects were conducted. One project had the objective of evaluating delta boun for outdoor measurements sites. The other project investigated delta boun for indoor environments, namely Hemi-anechoic rooms, and had the additional objective of investigating the relationship between room performance and other sound power measurement uncertainty components. For both studies the uncertainty was derived from identifying the limits of the measurement error resulting from the boundary reflections. In practice this involved constructing a 'worst-case' environment in which to conduct the sound power measurements.
Results
The summary of the the recommended A-weighted standard uncertainty value for delta boun resulting from the studies are as follows:
| For outdoor test site |
0.2 |
| For Hemi-anechoic rooms |
0.0 |
NPL measured a value for delta bound as large as 0.6 dB for some 1/3 octave bands, for both indoor and outdoor environments.
The recommended value of delta boun in hemianechoic rooms is only applicable to rooms that were qualified according to ISO 3745 using discrete tones from the test source. A room qualified using the procedure in ISO 3745:2003, using broad band noise, has been found to be inadequate and may result in a significantly larger uncertainty value.
Recommendations
A the recommendations from the research can be summarized as follows:
| The ISO 3745:2003 'inverse square' test has been found to be inadequate and should therefore not be performed using a broad-band test signal. |
Further information
The report detailing aspects of this research into evaluation of delta boun for hemi-anechoic rooms is available here: NPL Report_DQL_AC7 (1,268 kB)
delta angle
Results
This uncertainty component Allows for error resulting from angular difference between measurement surface normal and sound emission direction of the noise source.
Research Summary
Investigations of systematic errors associated with the shape of the measurement surface have been made. Consideration has been given to the differences between hemispherical and parallelepiped measurement arrays when determining sound power level. The effect of source characteristics has been examined, with variation of source location within the constraints of a range of delineated machine dimensions. It was considered appropriate to adopt an analytical approach for this study because this modeling approach enables a greater volume of data to be obtained than from measurement.
The summary of the the recommended A-weighted standard uncertainty values for delta Angle resulting from this study is as follows:
| For small sources |
0.4 |
| For medium size sources |
0.6 |
| For large sources |
1.0 |
delta met
This is the uncertainty allowing for any error resulting from variation in meteorological conditions
Research Summary
Annex E, of the proposed ISO 3744, provides an expression to normalize the measured Sound power level to Standard atmospheric temperature and pressure. This expression has been used to calculate sound power level corrections from a sample of local historical meteorological data. This preliminary study evaluates the statistical distribution of the corrections providing an estimate of delta met.
Results
A summary of the the calculated A-weighted standard uncertainty value for delta met, based on meteorological data between August 2003 and July 2004, for South west London, is as follows:
| Value for delta met |
0.1 |
Related NPL Machinery Noise Research
NPL has been conducting research into machinery noise measurement and its uncertainties through out the previous program (1998 to 2001). Reports are available below:
Assessment of sound power levels and associated measurement uncertainties of construction equipment
Sound power determinations for the Outdoor Machinery directive, using the procedures in ISO3744 and ISO4872, have been compared NPLReport_CMAM65.pdf (550 KB)
The establishment of a Rapid Method of Making Sound power Measurements of Machinery Noise Emission
Sound power determinations have been carried out using a variety of machines, in a variety of acoustical environments. The relationship between measurement accuracy, directivity index, number of microphone positions and k2has been investigated. NPLReport_CMAM19.pdf (58,758 KB)
An investigation into the measurement of the environmental correction factor used in the determination of the sound power level and emission sound pressure level of machines
Various methods of determination of Environmental correction factor, k2A, have been compared in test rooms with a variety of acoustical properties. This study concludes that the absolute comparison method, (using a RSS), provides the most consistent assessment of k2. NPLReport_CIRA(EXT)009 (226 kB)
Assessment of Reproducibility Uncertainties for use in International Standards on the Determination of Sound Power
Four Laboratories carried out measurements on three noise sources encompassing a wide range of directivity indices. The sound power levels were determined using several microphone arrays with different distributions on the hemispherical enveloping surface and reproducibility uncertainty calculated for each array. NPLReport_CMAM51.pdf (1,586 kB)
The Effect of Enveloping Surface Shape and Size on the Determination of Sound Power
This experimental study assess the effect of enveloping surface size and shape on the determination of Sound power level using ISO 3744 (1994) NPLReport_CMAM30 (8,491 kB)