First ideas for a revision of ISO 9614 Volker Wittstock 1 Physikalisch-Technische Bundesanstalt Bundesallee 100, 38116 Braunschweig, Germany Spyros Brezas 2 Department of Music Technology and Acoustics, Hellenic Mediterranean University E. Daskalaki, Perivolia, 74133, Rethymno, Greece Fabian Heisterkamp 3 Federal Institute for Occupational Safety and Health (BAuA) Friedrich-Henkel-Weg 1-25, 44149 Dortmund, Germany

ABSTRACT ISO 9614 describes the determination of sound power levels by the two microphone intensity technique. Despite some physical advantages the method is relatively seldom used. To improve this situation, theoretical and experimental investigations were performed, which focused mainly on the adequacy of the currently standardized indicators and criteria. This paper summarizes these results and develops first ideas for a revision of ISO 9614. One main proposal is to perform measurements generally in the one-third octave bands between 50 Hz and 10 kHz with a 12 mm spacer and to apply appropriate corrections in the highest three one-third octave bands. Furthermore, it is proposed to require additional measurements with a larger spacer at lower frequencies, if the dynamic capability of the instrument is insufficient. The A-weighted sound power level is calculated from the one-third octave band levels, which may have different grades of accuracy. Calculating different A-weighted sound power levels from different sets of frequency bands fulfilling criteria for different grades of accuracy then allows for designating a grade of accuracy to the A-weighted sound power level. This is considered to be a major improvement since the A-weighted sound power level is the main descriptor for the sound emission of sources.

1. INTRODUCTION

The intensity methods ISO 9614-1, -2 and -3 [1-3] are the most modern methods for the determination of the airborne sound power level. Their use is advantageous compared to the sound pressure methods under certain field conditions, such as high background noise or reflective environments. Furthermore, intensity methods can be applied in the near field of the source under

1 volker.wittstock@ptb.de 2 sbrezas@hmu.gr 3 heisterkamp.fabian@baua.bund.de

test. Nevertheless, intensity methods are relatively seldom used in practice due to several reasons. Besides the relatively high costs of the measurement equipment, the current standardization is often held responsible for this. To improve the applicability of the intensity method for sound power determination, a common research effort was undertaken by PTB and BAuA. The final report is available online [4] and in print upon request. One main result of this research is a proposal for a future revision of ISO 9614 (all parts). This contribution provides an overview of this proposal. It is not meant to present fixed solutions but to start a discussion on this topic.

2. MAIN PROPOSALS

The current ISO 9614 series comprises three parts. Whereas part one covers three grades of accuracy for measurements at discrete points, the scanning methods are described in parts two and three. Part two covers engineering und survey grade measurements, and part three precision grade measurements. This structure has developed historically and does not follow a logical structure. It is therefore proposed to reduce the 9614-series to two parts, one for measurements at discrete points and one for scanning measurements. Both parts should include all three grades of accuracy.

Another issue of the current ISO 9614 series are the indicators. Whereas the difference between the signed pressure - intensity indicator 𝐹 𝑝𝐼 n and the pressure - residual intensity index 𝛿 𝑝𝐼 0 is clearly linked to the measurement error caused by the phase mismatch between both measurement channels (see Figure 1), neither theoretical nor experimental proof could be found for the application of the unsigned pressure-intensity indicator 𝐹 𝑝|𝐼 n | and the field non-uniformity indicator 𝐹 S within the research project [4]. It thus seems logical to remove these indicators from a future version of the intensity standards which would be a significant simplification.

Figure 1: Difference between the sound power level measured by a p-p probe and the correct sound

power level Δ𝐿 𝑊,𝐼 as a function of the difference between the signed pressure - intensity indicator

𝐹 𝑝𝐼 n and the pressure - residual intensity index 𝛿 𝑝𝐼 0 for positive (solid line) and negative (dashed

line) phase mismatches The current ISO 9614 series limits the difference between the sound power level measured by a p-p probe and the correct sound power level Δ𝐿 𝑊,𝐼 by requiring 𝐹 𝑝𝐼 n −𝛿 𝑝𝐼 0 < −10 dB for precision and engineering grade, and 𝐹 𝑝𝐼 n −𝛿 𝑝𝐼 0 < −7 dB for survey grade measurements. On the base of Figure 1, it is now proposed to distinguish between all three grades of accuracy. If the grade of accuracy is linked to an uncertainty of the sound power level 𝑢(𝐿 𝑊 ) , it is considered sensible to use half of this uncertainty for the sound power error due to phase mismatch Δ𝐿 𝑊,𝐼 (see Equation (1)). From Figure 1, the criteria given in Table 1 can then be derived. The proposed criterion for precision grade

measurements is identical to the current one, whereas the criterion for engineering and survey grade measurements are loosened by 2 dB each.

|∆𝐿 𝑊,𝐼 | < 𝑢(𝐿 𝑊 )

2 (1)

Table 1: Uncertainty of the sound power level for different grades of accuracy and tolerable ∆𝐿 𝑊,𝐼

Grade of accuracy 𝒖(𝑳 𝑾 ) |∆𝑳 𝑾,𝑰 | 𝑭 𝒑𝑰 n −𝜹 𝒑𝑰 𝟎

Precision 1,0 dB < 0,5 dB < -10 dB

Engineering 1,5 dB < 0,7 dB < - 8 dB

Survey 4,0 dB < 1,7 dB < - 5 dB

The current ISO 9614-1 requires a minimum number of measurement points 𝑁 min , which is calculated from the field non-uniformity indicator 𝐹 S . From the results of the research project [4], an application of this criterion does not seem to be justified. Nevertheless, the derivation of a new criterion was also not successful. It is therefore proposed to use the recommendations for the initial measurements defined in the current ISO 9614-1, where the minimum number of measurement points is linked to the area of the enveloping surface 𝑆 , see Equation (2).

10 for 𝑆< 10 m 2

𝑆m ⁄ 2 for 10 m 2 < 𝑆< 50 m 2

𝑁 min = {

, (2)

50 for 𝑆> 50 m 2

The currently standardized procedures require a qualification of the measurement results by checking several criteria. One possible outcome of this procedure is that no sound power level result is obtained which usually happens in only few one-third octave bands. To avoid such an unpleasant outcome of a measurement it is proposed to use the measured sound pressure level without the environmental correction 𝐾 2 to calculate an upper limit of the sound power level in the respective frequency bands.

The upper frequency limit of the current ISO 9614-series is 6,3 kHz. This is linked to the finite difference error caused by the spacing between the microphones. Unfortunately, microphone spacing is not clearly addressed in the current ISO 9614-series. It is therefore proposed to include information on the effect of microphone spacing on the finite difference error, but also on the pressure - residual intensity index 𝛿 𝑝𝐼 0 . To cover the frequency range usually applied for A- weighted sound power levels, it is proposed to extend the frequency range up to 10 kHz. This requires a default correction to be given in the standard to handle the finite difference error. For a 12 mm spacer, such a correction is proposed. At the same time, probe manufacturers are asked to provide individual corrections for their probes.

A final issue are low noise sound sources. Due to the comparatively low sensitivity of probe microphones, it can be more difficult to determine the sound power of low noise sound sources by intensity methods than by sound pressure methods due to electrical background noise. This aspect is not addressed in today’s ISO 9614-series. It is therefore proposed to include a note with an appropriate recommendation, e.g. that the measured sound pressure level should be 10 dB above the electrical background noise level.

3. PROPOSED MEASUREMENT STRATEGY

The proposed measurement strategy is focused to obtain a certain grade of accuracy for the A- weighted sound power level. The main workflow is as follows: 1. perform sound pressure and sound intensity measurements at 𝑁 points on the enveloping surface

or perform scans on the 𝑁 partial surfaces, use a 12 mm spacer and include all one-third octave bands between 50 Hz and 10 kHz 2. calculate at first the indicator 𝐹 𝑝𝐼 n , and then the sound power level (including the finite

difference correction for frequencies starting from 6,3 kHz) for all one-third octave bands fulfilling the three following criteria

𝐹 𝑝𝐼 n −𝛿 𝑝𝐼 0 < −𝐾 (3)

𝐿 𝑝 ̅̅̅ > 𝐿 𝐼 n ̅̅̅̅ – 2 dB (4)

𝑁

1 𝑁 ∑sgn (𝐼 n,𝑖 )10 0.1 𝐿 𝐼n,𝑖 /dB

> 0 (5)

𝑖=1

where Criterion (5) ensures that the surface-averaged intensity is positive and 𝐾 is chosen from

10 dB for precision grade 8 dB for engineering grade

𝐾= {

(6)

5 dB for survey grade

3. for one-third octave bands without a qualified result, use the measured sound pressure levels to

estimate an upper limit for the sound power level 4. calculate the A-weighted sound power level: a) from all one-third octave bands ( 𝐿 𝑊A ) and b)

from only those one-third octave bands, which qualified for the desired grade of accuracy ( 𝐿 𝑊A,BL ) 5. 𝐿 𝑊A is qualified with the desired grade of accuracy when

𝐿 𝑊𝐴 −𝐿 𝑊A,BL < 𝐶 A (7)

0,5 dB for precision grade 0,8 dB for engineering grade

𝐶 A = {

(8)

2,0 dB for survey grade

where 𝐶 A is proposed to be half of the uncertainty of the sound power level. When criterion (7) is not fulfilled, it is recommended to reperform the measurements with changed parameters such as measurement distance or microphone spacer.

4. CONCLUSIONS

Standards for the determination of sound power levels by measuring intensity should be revised. One main aim of this revision should be a major simplification of the standards to improve their applicability. This contribution is meant to initiate a discussion on a revision of the ISO 9614-series.

A simplified 9614-series would help machinery manufacturers and other users by reducing the measurement effort for the sound power determination. This, in turn, will improve the reliability of noise emission data of machines and ensure a fair competition towards quieter machines. Reliable noise emission data help employers and users of machines to preferably select and buy quieter machines (Sell and Buy Quiet). This way, employers can better protect their workers from noise. 5. ACKNOWLEDGEMENTS

Presented results were obtained within a research project undertaken at Physikalisch-Technische Bundesanstalt (PTB) and funded by the Federal Institute for Occupational Safety and Health (BAuA) as a subproject (F2450) of a focus project aiming at the simplification of noise emission measurement methods. 6. REFERENCES

1. ISO 9614-1:1993 Acoustics — Determination of sound power levels of noise sources using

sound intensity — Part 1: Measurement at discrete points 2. ISO 9614-2:1996 Acoustics — Determination of sound power levels of noise sources using

sound intensity — Part 2: Measurement by scanning 3. ISO 9614-3:2002 Acoustics — Determination of sound power levels of noise sources using

sound intensity — Part 3: Precision method for measurement by scanning 4. S. Brezas, F. Heisterkamp, and V. Wittstock. Practice-oriented simplification of noise emission

measurement methods, subproject 2: Practical adaptation of the sound intensity measurement method. Report, Federal Institute of Occupational Health and Safety (BAuA), doi:10.21934/baua:report20210914 (online), 2021.