Adventures in uncertainty: A former consultant’s view of manufacture-driven acoustical testing

Abstract

Acoustical consultants commonly rely on laboratory testing of wall and floor/ceiling assemblies to inform their designs and to evaluate noise control solutions. Today, much of the available laboratory testing comes from manufacturers of noise control products. Laboratory tests are taken as indicative of assembly and product performance, but there is significant uncertainty inherent in the ASTM E90 and E492 test methods. When evaluating test results to design assemblies or compare products, this uncertainty is often given insufficient attention and test results are taken as de facto indicators of assembly performance. Manufacturers are faced with how to handle this uncertainty when deciding what to test, where to test, how to test, and ultimately what results to make public. Manufacturers are often accused of “cherry-picking” test data, but the reality is often more complicated than that. In this presentation, the author will discuss his experience working at product manufacturers and deciding what to test and what data to make public. The presentation will also look at some examples of how the uncertainty in laboratory testing could affect the outcome of the tests. Finally, the author will discuss other ways of presenting test data that aim to reduce some of the uncertainty and make it easier for acousticians to make good design decisions.