Audio Engineering Society Standards Committee

December 2001 report of SC-04-03

[Last printing 30 April 2002] Report of the SC-04-03 Working Group on Loudspeaker Measurement and Modeling of the SC-04 Subcommittee on Acoustics meeting, held in conjunction with the AES 111th Convention in New York City, NY, US, 2001-12-02.

Chair David Prince convened the meeting. The agenda and the report from the previous meeting at the AES 109th Convention were approved as written.

Current development projects

AES-1id-R Review of information document, AES-1id-1991 (r1997) Plane-Wave Tubes: Design and Practice

Marshall Buck agreed to review the document. Any other contributors are encouraged to contact the reflector. The target is a status report for 2002-05.

AES-5id-R Review of AES-5id-1997 AES information document for Room acoustics and sound-reinforcement systems -- Loudspeaker modeling and measurement -- Frequency and angular resolution for measuring, presenting and predicting loudspeaker polar data

No action was taken.

AES2-R Revision of AES2-1984 (r1997) Specification of loudspeaker components used in professional audio and sound reinforcement

There was no report. The new target for a PWD is 2002-05.

AES19-R Review of AES19-1992 (r1998) AES-ALMA Standard Test Method for Audio Engineering -- Measurement of the lowest resonance frequency of loudspeaker cones

No action was taken.

AES-X72 Acoustic Center of Loudspeakers

No report was ready.

AES-X103 Large Signal Parameters of Low-Frequency Loudspeaker Drivers

Extensive reports were given by W. Klippel and A. Voishvillo. Papers summarizing the contents of those reports are on the FTP site.

Klippel reported:

1) the group needs to review measurements of non-linear motor parameters and agree which form o be used in presentation, then decide how they can be compared;

2) in determining peak displacements (X), the group needs to consider the limiting factor (XBl, Xle, Xc);

3) the group needs to sum up to global Xmax; fortunately, symptom based and parameter based analyses are closely related.

A discussion followed.

D. Gunness would like to have plots rather than single number since a rounded flux density times effective winding length (Bl) curve versus X curve has different sound than a squared off Bl curve. A 50 % number is a good indication of area or total energy under the curve.

Geddes agreed that a definition should be as simple as possible but he felt that we do not have enough information yet. Mevertheless, Klippel's proposal may be adequate for now.

Gunness thinks we should move away from the Xmax as a term (to get away from past definitions). He would like to know what distortion level correlates with 50 % point on the Bl curve. S. Temme agreed that one needs to see curves rather than single numbers. M. Buck agreed not to call it Xmax. Voishvillo commented that only Xc is really pertinent to destruction of driver.

Klippel noted that his report differentiates between criteria for physical limits and criteria for perceptual limits. Prince asked is it possible to keep these criteria simple enough for wide interpretation by both sophisticated and unsophisticated users. Geddes pointed out that because of the dominance of the application conditions, for example, the enclosure, there is no simplification, so the end user must become educated that it is a complicated relationship.

J. Brown noted he would like to see families of curves rather than single numbers. Gunness countered that inevitably, someone is going to create a table and not use the curves so it is worthwhile to have a set definition of the single number.

D. Clark pointed out that there is a need for standardization of Xmax in the automotive industry and he would like to see it weighted from XBl.

Voishvillo's presentation followed. in the discussion M. Kleiner noted that he has used multitone a lot with a very efficient system that works well. Klippel commented that a coherence/incoherence ratio requires huge averaging time. Geddes disagreed about the averaging time but agreed that it should not be used. Buck pointed out that coherence also requires anechoic measurements. Temme suggested the need to talk about measured numbers but also the need need for psychoacoustical considerations. This suggestion led into discussions regarding project AES-X129.

Klippel again asked for input leading to Xmax conclusions. Gunness asked for a conclusion about whether Xmax can be derived by parameter measures, for example, 70 % on the BL curve; or performance measures, for example, 10 % distortion. Klippel preferes to work on parameter-based measurement but stressed more importantly that the existing common definition be abandoned. Prince pointed out that the AES2-R discussions have dispensed with the old measurement and instead refer to IEC 60268-5.

AES-X129 Loudspeaker Distortion Perception and Measurement

Prince reintroduced notes previously released to group prior to 2001-05. Geddes corrected the notes in that his initial concept was to use the Internet for large scale testing, but concluded that it is not feasible.

Kleiner noted that Bose had used society meetings to validate its modeling systems. Geddes ackowledged that the study is needed to move forward but felt that funding would not be available. Prince suggested that a task group might be able to coordinate enough volunteer effort to move forward. T. Welti indicated that Harman may have some of the material and software to start. Clark and others noted that many of the needed signals can be produced from available equipment such as Klippel's system.

New projects

No project requests were received or introduced.

New Business

There was no new business

The next meeting is scheduled to be held in conjunction with the AES 112th Convention in Munich, Germany.

For more information about standards activity:

Back to AESSC Meeting Reports

(C) 2002, Audio Engineering Society, Inc.