Simplification of ISO 3744 – making the most important standard for determination of sound power levels easier to use Fabian Heisterkamp 1 Federal Institute for Occupational Safety and Health (BAuA) Friedrich-Henkel-Weg 1-25, 44149, Dortmund, Germany Jeff Schmitt 2 VIacoustics Austin, TX., USA Joey Hook 3 VIacoustics Austin, TX., USA

ABSTRACT Sound power level ( 𝐿 𝑊 ) is the key quantity to describe the noise emission of products and is needed to reduce noise at work, at home, and in the environment, e.g. for promoting and selecting low noise products (Sell and Buy Quiet). ISO 3744:2010 is the main standard referred to by most product test codes and regulatory requirements for determining L W . It is used by laboratories, noise control en- gineers, and manufacturers. Due to its structure and length, the usability of the current document is limited. The main body of ISO 3744:2010 contains so much information that the basic concept of sound power is obscured. To improve usability of ISO 3744:2010 three projects were launched: 1. A revision and restructuring of ISO 3744 with a simplified main body and a series of Annexes for special cases; 2. A new standard: ISO/CD 5114-1 on the uncertainty of 𝐿 𝑊 determined from sound pressure levels; 3. A new standard: ISO/CD 26101-2 to determine the environmental correction used in determining L W . These new standards contain specific information needed by some users while ISO/CD 3744 focuses on determining 𝐿 𝑊 . We present key aspects of all three committee drafts and explain how and why ISO 3744 should be revised. 1. INTRODUCTION

The sound power level 𝐿 𝑊 is a measure of the sound energy emitted by a product over time. The sound power level is independent of the environment, which makes it an invaluable tool for reducing noise at work, at home and in the environment, e.g. by promoting and enabling the selection of low noise products in accordance with the Sell and Buy Quiet concept [1].

The sound power level is used in many regulations, product specifications and standards. For ex- ample, the following regulations require the determination of the 𝐿 𝑊 for some or all the products in

1 heisterkamp.fabian@baua.bund.de

2 jeffs@viacoustics.com

3 joey@viacoustics.com

worm 2022

their scope: The EU Outdoor Noise Directive [2], the EU Machinery Directive [3], Australian regu- lations regarding noise labeling [4, 5], and the EU Ecodesign Directive [6].

ISO 3744 [7] is the standard for the determination of the sound power level, which is referred to by most regulations, machine-specific noise test codes or safety standards, product specifications and standards. It is the most used standard for 𝐿 𝑊 determination, because it is suitable for measurements indoors and outdoors and can be used not only in hemi-anechoic test rooms, but also in non-ideal acoustic environments, e.g. industrial halls, workshops, etc. Still, it delivers results of accuracy grade 2 (engineering grade), see ISO 12001 [8].

ISO 3744 is one standard of the group ISO 3740 to ISO 3747 that deal with the determination of sound power levels of noise sources and are all successors of ISO/R 495:1966 [9]. Here, ISO 3740 [10] is a guide to select the appropriate standard to determine 𝐿 𝑊 and as such is not limited to ISO 3741 to 3747 that cover the 𝐿 𝑊 determination from sound pressure level measurements.

The current version ISO 3744:2010 is a revision and merger of ISO 3744:1994 [11], which covered an engineering method to determine the 𝐿 𝑊 of noise sources using sound pressure measurements in an essentially free field over a reflecting plane, and ISO 4872:1978 [12]. This standard specifically dealt with the measurement of noise emissions of construction equipment and was a method for test- ing against a compliance limit. Furthermore, the determination of sound energy levels was added to ISO 3744 during that revision.

Thus, the current version extended the scope of the standard, but this expansion came with a price regarding its usability (see Sec. 2). Sec. 3.1 describes how ISO 3744 has been simplified and restruc- tured in the current revision proposal, ISO/CD 3744:2021 [13]. Sec. 3.2 justifies why parts of the determination of the measurement uncertainty were moved to the new ISO/CD 5114-1:2021 [14] and discusses the structure and advantages of the new standard. Another standard supporting the revision of ISO 3744:2010 is covered in Sec. 3.3. The new ISO/CD 26101-2:2021 [15] contains certain meth- ods to determine the environmental correction, so that the revision proposal for ISO 3744 only con- tains the preferred method according to ISO 3744:2010 A.1 [7], which is the absolute comparison test. Sec. 4 concludes the paper by summing up the key aspects of the revision proposal and its ad- vantages.

2. WHY ISO 3744:2010 NEEDS TO BE REVISED AND RESTRUCTURED

As already hinted at in Sec. 1, the last revision of ISO 3744 extended the scope of the standard, but at the same time reduced its usability. Feedback from typical users of the ISO 3744 test method, which are often engineers without a background in acoustics, was that the 2010 version was too hard for them to understand.

While ISO 3744:1994 has a length of 31 pages, the current ISO 3744:2010, although maybe in part due to a different formatting, features 78 pages. The number of definitions increased from 18 to 25 mainly due to the introduction of the sound energy level. Note that the new definitions regarding the sound energy level are very similar to those of the sound pressure level and sound power level, respectively.

The introduction of sound energy levels did not only increase the number of definitions, it also increased the number of equations by seven (count based solely on ISO 3744:2010) and caused an increase of the length of the main body due to a duplicate sub clause, i.e. “8.3 Determination of sound energy levels “, that is very similar to that for the sound power level. Furthermore, but this is the least problematic aspect, it leads to many “duplicate” symbols, e.g. ISO 3744:2010 Cl. 4.2.3 “ 𝐿 𝑊 or 𝐿 𝐽 ” and “ 𝐿 𝑊A or 𝐿 𝐽A ”, Cl. 9.1 “uncertainties of sound power levels, 𝑢(𝐿 𝑊 ) , in decibels, and sound energy levels, 𝑢(𝐿 𝐽 ) , in decibels,…”, etc.

worm 2022

To the knowledge of the authors and according to ISO 3740 Clause 1 Note 2 “The sound energy level mentioned in the ISO 3740 series of standards is not addressed in this document as it is not mentioned in any legal requirement and therefore not used in practical measurement. […]” [10], sound energy levels are not part of any regulatory requirement and thus hardly used in practice.

Furthermore, many special cases are covered in the document and specific information needed only by a few users is included, sometimes even in the main body of the standard. For example, Clause “9 Measurement uncertainty” has a length of two pages, although a special Annex “Annex H Guidelines on the development of information on measurement uncertainty” provides further details on this topic on 11 pages. Many of these details, mainly regarding the modeling approach according to the Guide to the expression of uncertainty in measurement (GUM) [16], are not needed by many users of ISO 3744.

Another example of the inclusion of special cases and in this case even less accurate methods is the determination of the environmental correction. This is covered in Annex A of ISO 3744:2010. In the current version (2010), five different methods are described. As a result, Annex A has a length of 5 pages and, more importantly, contains methods that are less reliable, especially in typical industrial environments [17, 18]. Especially, the method based on an estimation of the equivalent sound ab- sorption area 𝐴 of the room (see ISO 3744:2010 A.3.5) is not acceptable for an engineering grade standard, even if it is only allowed for the determination of an A-weighted sound power level. On the other hand, the same Annex A refers to the absolute comparison test as “the preferred method”. This method is suitable for all users, who own a reference sound source, and can be used indoors and outdoors to qualify the acoustic environment and determine the environmental correction 𝐾 2 .

Due to the last revision, the document has a very long main body, i.e. a very long section that every user of the document must read, even if they do not need many parts of the text. The main body of ISO 3744:1994 has 13 pages, while that of the current version has a length of 31 pages. Thus, the basic concept to determine the sound power level using sound pressure has been obscured by the last revision. The following section describes how we propose to address this problem, restructure and revise ISO 3744:2010, so that its main body will focus on the determination of sound power levels.

3. THE SIMPLIFICATION OF ISO 3744:2010 AND SUPPORTING STANDARDS

3.1. How ISO/CD 3744:2021 simplifies and restructures ISO 3744:2010 The key concepts associated with the revision, simplification and restructuring of ISO 3744 in ISO/CD 3744:2021 [13] are as follows:

 Focus the main body of the document on the basic concept of sound power level determina-

tion in an essentially free field over a reflective plane under typical measurement conditions. That being the concept that sound power level is determined from:

o The average sound pressure level 𝐿 𝑝 തതത measured on a surface that envelopes the device under test, plus… o A term that corrects for the influence of background noise ( 𝐾 1 ), plus… o A term that corrects for the influence of the test environment ( 𝐾 2 ), plus… o A term that accounts for the surface area of the measurement ( 10 lg(𝑆𝑆 0 Τ ) )  Move most of the special conditions related to sound power level determination to the An-

nexes, thus making the main body shorter and more concise.  Move all the special cases related to the qualification of the test environment and determina-

tion of the 𝐾 2 term to a new standard ISO/CD 26101-2:2021, retaining only the preferred absolute comparison test using a calibrated Reference Sound Source in Annex A of ISO/CD 3744:2021.

worm 2022

 Move all the details and mathematical modeling related to measurement uncertainty, other

than the stated standard deviation of reproducibility ( 𝜎 R0 ) and the standard deviation of op- erating and mounting conditions ( 𝜎 omc ), to a new standard ISO/CD 5114:2021 [14].  Removal of the concept of sound energy level from ISO/CD 3744:2021. A reference to ISO

3744:2010 for the determination of sound energy levels is being included until Working Group 28 can find a project leader to develop a standalone sound energy level standard that would incorporate both the free field and reverberant field methods.

This restructuring has resulted in a main body of ISO/CD 3744:2021 that consists of 21 pages that outline the basic concepts of sound power level determination. The Definitions section now contains 20 defined terms. There are now nine (9) Annexes to ISO/CD 3744:2021.

There is one new Annex in ISO/CD 3744:2021: Annex I on Laboratory Procedures for Reduc- tion of Uncertainties Associated with Sound Power Level Determinations has been added. The 2010 revision of ISO 3744 was based on research that indicated the sound power levels could be deter- mined with a standard deviation of reproducibility of 1,5 dB(A) in test environments with 𝐾 2A cor- rections as large as 4 dB.

However, ISO 3744 is also used by many laboratories that conduct testing in hemi-anechoic test facilities that have been qualified as precision grade test environments. Annex I of ISO/CD 3744:2021 recognizes that sound power level determinations conducted in such laboratory test envi- ronments and using other best practices in acoustic measurement under the control of an ISO 17025 quality system will result in lower standard deviations of reproducibility. This reduced measure- ment uncertainty associated with Annex I can allow manufacturers to provide lower product noise declaration values with a high confidence that the measurements made to determine the declared value were made with reduced measurement uncertainty.

Other changes of note in ISO/CD 3744:2021 include the following:

 Elimination of the absolute criterion for background noise that was added to the 2010 re-

vision of ISO 3744. It has been replaced with provisions that allow for statement of full compliance with the test method provided background noise levels in the test environ- ment used are low enough to validate a regulatory, statutory or declared value with at least 6 dB of signal to noise ratio relative to the criteria being validated.  Incorporating portions of ISO TC 43/SC 1 Guide 1 on drafting instrumentation require-

ments to include requirements for more modern modular, computer-based instrumenta- tion systems that are being used by many laboratories and acoustic consultants for meas- urements.  Moving special cases of measurement surfaces adjacent to two or more reflecting planes

into the Annexes on hemispherical and parallelepiped measurement surfaces.

3.2. Measurement uncertainty in the determination of sound power levels using sound pres- sure – ISO/CD 5114-1:2021 The potential of this document [14] to help simplify other standards goes beyond the revision of ISO 3744. The idea is to have a separate standard for the measurement uncertainty in the determina- tion of sound power levels using sound pressure, so that duplicate, but slightly different information on measurement uncertainty, e.g. in ISO 3744:2010 [7] and in ISO 3745 [19], is avoided. Further- more, ISO/CD 5114-1 contains more background information on measurement uncertainty and the GUM [16] than the existing Annexes on uncertainty in the ISO 3741 to ISO 3747 group of stand- ards.

worm 2022

ISO/CD 5114-1:2021 (see Clause 4) starts with the general concept to determine the uncertainty of measured sound power levels based on the GUM [16]. Here, the total standard deviation 𝜎 tot is introduced and how it can be determined from the standard deviation of reproducibility 𝜎 R0 and the standard deviation of the operating and mounting conditions 𝜎 omc .

The determination of the latter is described in Clause 5. Here, the standard also explains that in- vestigations of this quantity should have priority, because the contribution of 𝜎 omc to the total standard deviation 𝜎 tot is usually larger than that of 𝜎 R0 .

Regarding the standard deviation of reproducibility 𝜎 R0 , Clauses 6 and 7, describe the two ap- proaches to determine it according to the GUM:

1. The determination of this quantity by round robin tests. 2. From a detailed uncertainty budget using the modeling approach. The main body of the standard concludes with the determination of the total standard deviation 𝜎 tot and provides examples based on the standard deviation of reproducibility 𝜎 R0 assigned to dif- ferent accuracy grades (see ISO 12001:1996 [8]) and different values for the standard deviation of the operating and mounting conditions 𝜎 omc .

The Annexes of ISO/CD 5114-1:2021 [14] contain specific models to determine 𝜎 R0 for the dif- ferent concepts to determine the sound power level from sound pressure:

1. In (approximated) free fields using the direct enveloping method (Annex A) according to

ISO 3744, ISO 3745 or ISO 3746 2. In (approximated) diffuse fields using the direct enveloping method (Annex B) according to

ISO 3741 3. Using a reference sound source, i.e. according to ISO 3743-1, ISO 3743-2 or ISO 3747 The new standard uses the opportunity to slightly revise and extend these models and, even more important, unifies and summarizes the models from the different standards of the group from ISO 3741 to ISO 3747. This will allow future revisions of other standards of this group to refer to this document and simplify the information on measurement uncertainty in the standards themselves.

3.3. Determination of the environmental correction – ISO/CD 26101-2:2021 This standard [15] collects the methods for determining the environmental correction 𝐾 2 that is re- quired when using ISO 3744 in a non-ideal acoustic environment outside of a hemi-anechoic cham- ber, or when using ISO 3746 to determine the sound power level.

Many of the methods that are part of Annex A of ISO 3744:2010 [7] are not used by most users. For example, laboratories operating their own hemi-anechoic chamber and testing products accord- ing to ISO 3744 do not need to use any method contained in Annex A of ISO 3744:2010. Furthermore, Arendt et al. demonstrated that in typical industrial environments only the absolute comparison test, see ISO 3744:2010 A.2, yields reliable results for the environmental correction 𝐾 2 , while the methods based on room absorption, see ISO 3744:2010 A.3 [7], and especially the estima- tion method, see ISO 3744:2010 A.3.5, were less reliable in these environments [17,18]. In [17], it is recommended that only the absolute comparison test should be used for ISO 3744.

To simplify ISO 3744, only the absolute comparison test remains in that standard, while ISO/CD 26101-2:2021 contains all the methods from Annex A of ISO 3744:2010. Furthermore, ISO/CD 26101-2:2021 features a method to qualify a hemi-anechoic room for parallelepiped and cylindrical measurement surfaces. This is intended for laboratories testing products according to ISO 3744 and will ensure that the qualified volume for these measurement surfaces is maximized compared to a

worm 2022

qualification according to ISO 26101-1 [19] and ISO 3745 [20], which is intended to qualify a vol- ume in a hemi-anechoic chamber for hemi-spherical measurement surfaces only (see also [21]).

ISO/CD 26101-2:2021 [15] starts with a description of the different kinds of qualification proce- dures, four procedures corresponding to six methods, and a flowchart guiding the selection of the proper method to determine 𝐾 2 . It continues with the absolute comparison test (ISO/CD 26101- 2:2021 Clause 5) that is kept identical to the one in ISO/CD 3744, i.e. the revision proposal for ISO 3744.

This is followed by the qualification procedure based on room absorption that encompasses three methods: The direct method using a reference sound source (RSS) to determine the equivalent sound absorption area, 𝐴 , of the room; the two-surface method, where average sound pressure levels on two different measurement surfaces (ratio of the surface areas S at least 2) are used to calculate 𝐴 ; and the reverberation method, where 𝐴 is calculated from the measured reverberation time.

These methods assume that the room has an approximately cubic shape and that the sound absorp- tion occurs on the walls and ceiling only. In practice, however, these conditions are often not met. Then, e.g. in typical industrial buildings, the sound propagation might be distorted and these methods might be less reliable than the absolute comparison test.

ISO/CD 26101-2:2021 Clause 7 describes how to qualify hemi-anechoic chambers for parallele- piped and cylindrical measurement surfaces. This will support laboratories with a hemi-anechoic chamber that perform measurements according to ISO 3744. A qualification according to ISO 26101- 2 helps them to maximize the qualified volume for these measurement surfaces.

For completeness and to keep it for reference from a future revision of ISO 3746:2010 [22], Clause 8 of the standard contains the approximate method based on an estimation of the equivalent sound absorption area of the room. This method is currently standardized in ISO 3744:2010 A.3.5 and ISO 3746:2010 Annex A.

All in all, ISO/CD 26101-2:2021 serves to remove special use cases and less favorable methods from ISO/CD 3744:2021. However, it is not limited solely to that purpose. Some methods have been revised and one method, the approximate method based on an estimation of the equivalent sound absorption area of the room, is no longer allowed for engineering grade standards such as ISO/CD 3744:2021. Furthermore, the inverse square law qualification of hemi-anechoic chambers for paral- lelepiped and cylindrical measurement surfaces fills the gap left by the existing standards [19, 20] for this qualification, because they only allow for a qualification of hemispherical measurement sur- faces/for hemispherical qualified volumes.

4. CONCLUSIONS

We describe how ISO 3744:2010 can be revised to improve its usability and make it easier to un- derstand. The relocation of special cases and details only needed by a small subset of users results in a condensed main body (21 pages instead of the current 31 pages). This revision will not only help to increase the usability of the most important sound power standard, but it can also serve as the blueprint to revise the whole ISO 3740 set of standards (ISO 3741 to ISO 3747).

Users that are new to acoustics or not mainly concerned with acoustics will benefit from the revi- sion of ISO 3744. Even experienced users/acousticians will benefit from the revised structure and simplified criteria regarding additional microphone positions and background noise, etc. Moreover, labs operating hemi-anechoic chambers can follow the procedures in Annex I of ISO/CD 3744:2021 to reduce the measurement uncertainty.

To keep most special use cases and the existing information for reference, two supporting stand- ards are introduced: While ISO/CD 26101-2 mainly supports the revision of ISO 3744 (and maybe

worm 2022

ISO 3746 in the future), ISO/CD 5114-1 can help to revise all standards of the ISO 3740 series. It contains a unified and revised version of the existing information on measurement uncertainty of sound power levels determined from sound pressure level measurements.

All in all, the existing knowledge of ISO 3744:2010 will be preserved in the course of the revi- sion, while making it more accessible and more understandable for users. This will help profes- sional users as well as small and medium-sized enterprises (SMEs) to determine the noise emissions of machines and other products more easily.

This, in turn, will improve the reliability of noise emission data of machines in general and en- sure a fair competition towards quieter machines. Reliable noise emission data will also help em- ployers to choose quieter machines more easily, so that they can better protect their workers from noise hazards. 5. ACKNOWLEDGEMENTS

We gratefully acknowledge the support and contributions of the members of ISO TC 43 “Acous- tics”/SC 1 “Noise”/WG 28 “Basic machinery noise emission standards”, especially Robert D. Hell- weg (convener of WG 28),Volker Wittstock (project leader for ISO 5114-1) and Kohei Shimoda (inverse square law qualification for parallelepiped and cylindrical measurement surfaces). 6. REFERENCES

1. Heisterkamp, F., Bengtsson Ryberg, J., Jacques, J., Verdaasdonk, A. Sell and Buy Quiet - the

extended concept to reduce noise (at work and at home). INTER-NOISE and NOISE-CON Con- gress and Conference Proceedings , 263(4) , 2011-2019, 2021. 2. European Parliament and European Council, Directive 2000/14/EC of the European Parliament

and of the Council of 8 May 2000 on the approximation of the laws of the Member States relating to the noise emission in the environment by equipment for use outdoors. , Official Journal of the European Communities , L 162 , 1-78, 2000. 3. European Parliament and European Council, DIRECTIVE 2006/42/EC OF THE EUROPEAN

PARLIAMENT AND OF THE COUNCIL of 17 May 2006 on machinery, and amending Di- rective 95/16/EC (recast), Official Journal of the European Union , L 157 , 24-86, 2006. 4. Government of Western Australia, Noise Abatement (Noise Labelling of Equipment) Regulations

(No. 2) 1985. Environmental Protection Act 1986 , 2549-2551, 1985. 5. State of New South Wales and Environment Protection Authority, The Protection of the Environ-

ment Operations (Noise Control) Regulation 2017 - Approved Methods for Testing Noise Emis- sions. ISBN 978-1-76039-353-3, 2017. 6. European Parliament and European Council, DIRECTIVE 2009/125/EC OF THE EUROPEAN

PARLIAMENT AND OF THE COUNCIL of 21 October 2009 establishing a framework for the setting of ecodesign requirements for energy-related products. Official Journal of the European Union , L 285 , 10-35, 2009. 7. ISO 3744:2010, Acoustics — Determination of sound power levels and sound energy levels of

noise sources using sound pressure — Engineering methods for an essentially free field over a reflecting plane. International Organization for Standardization, Geneva, Switzerland, 2010. 8. ISO 12001:1996, Acoustics — Noise emitted by machinery and equipment — Rules for the draft-

ing and presentation of noise test code. International Organization for Standardization, Geneva, Switzerland, 1996.

worm 2022

9. ISO/R 495:1966, GENERAL REQUIREMENTS FOR THE PREPARATION OF TEST CODES

FOR MEASURING THE NOISE EMITTED BY MACHINES. International Organization for Standardization, Geneva, Switzerland, 1966. 10. ISO 3740:2019, Acoustics — Determination of sound power levels of noise sources — Guide-

lines for the use of basic standards. International Organization for Standardization, Geneva, Swit- zerland, 2019. 11. ISO 3744:1994, Acoustics — Determination of sound power levels of noise sources using sound

pressure — Engineering method in an essentially free field over a reflecting plane. International Organization for Standardization, Geneva, Switzerland, 1994. 12. ISO 4872:1978, Acoustics — Measurement of airborne noise emitted by construction equipment

intended for outdoor use — Method for determining compliance with noise limits. International Organization for Standardization, Geneva, Switzerland, 1978. 13. ISO/CD 3744, Acoustics - Determination of sound power levels of noise sources using sound

pressure — Engineering methods for an essentially free field over a reflecting plane. International Organization for Standardization, Geneva, Switzerland, 2021. 14. ISO/CD 5114-1, Acoustics — Determination of uncertainties associated with sound emission

measures — Part 1: Sound power levels determined from sound pressure measurements. Interna- tional Organization for Standardization, Geneva, Switzerland, 2021. 15. ISO/CD 26101-2, Acoustics — Test methods for the qualification of the acoustic environment —

Part 2: Determination of the environmental correction. International Organization for Standardi- zation, Geneva, Switzerland, 2021. 16. ISO/IEC GUIDE 98-3:2008, Part 3: Guide to the expression of uncertainty in measurement

(GUM:1995). International Organization for Standardization, Geneva, Switzerland, 2008. 17. Arendt, I., Kurtz, P. Simplification of measurement procedures as a prerequisite for the provision

of reliable noise emission declarations - several problems regarding to ISO 3744. INTER-NOISE and NOISE-CON Congress and Conference Proceedings , 255(6) , 1987-1995, 2017. 18. Arendt, I., Kurtz, P., Verfahren zur Ermittlung der Umgebungskorrektur nach DIN EN ISO 3744

bei Ermittlung des Schallleistungspegels von Maschinen in industriellen Umgebungen. Lärmbe- kämpfung (03/2017) , 86-91, 2017. 19. ISO 3745:2012, Acoustics — Determination of sound power levels and sound energy levels of

noise sources using sound pressure — Precision methods for anechoic rooms and hemi-anechoic rooms. International Organization for Standardization, Geneva, Switzerland, 2012. 20. ISO 26101-2:2021, Acoustics — Test methods for the qualification of the acoustic environment

— Part 1: Qualification of free-field environments. International Organization for Standardiza- tion, Geneva, Switzerland, 2021. 21. Shimoda, K. Method of Inverse-Square Law Qualification of Hemi-Anechoic Chambers for Par-

allelepiped Measurement Surfaces in ISO 3744. INTER-NOISE and NOISE-CON Congress and Conference Proceedings , 254(2) , 469-476, 2017. 22. ISO 3746:2010, Acoustics — Determination of sound power levels and sound energy levels of

noise sources using sound pressure — Survey method using an enveloping measurement surface over a reflecting plane. International Organization for Standardization, Geneva, Switzerland, 2010.

worm 2022