EN ISO 3744, and how a decibel is manufactured

What the standard pins down

EN ISO 3744 is a method for determining the sound power level of a source, at what its own family calls engineering grade — one step below the precision methods that need an anechoic or reverberation room, and one step above the survey methods that will accept almost any site.

Sound power cannot be measured directly. What a microphone reads is pressure at a point, and the method’s job is to turn a set of those readings into a property of the source. It does that by fixing a measurement surface — a hemisphere, or a box, enveloping the machine over a reflecting plane — and specifying where on that surface the microphones go, how many there must be, and how their readings are combined.

It then fixes the corrections. A test site is never acoustically perfect, and the standard defines how the environment’s contribution is determined and subtracted, and how much background noise is tolerable before the result stops meaning anything. It sets out how long each measurement runs and how the machine’s operating condition must be recorded alongside the number.

And it fixes what must be reported. A sound power level stated without the method, the operating condition and the uncertainty is an incomplete result under this document, even when the arithmetic is correct.

What it does not promise

It does not say what the machine should be doing while it is measured. That is deliberate: a document that fixed the operating condition would only suit one kind of machine. For a lawnmower the condition comes from the mower standard, for a shredder from the shredder standard. This is why a sound power level obtained under this method for two different kinds of machine is not a like-for-like comparison.

It does not produce the figure on the housing. What it produces is a measured level. The guaranteed level that gets marked is that value with an uncertainty allowance added and a round-up applied, and it belongs to the noise directive rather than here.

It does not describe annoyance. A-weighting approximates the ear’s response at moderate levels, and it flattens the difference between a whine and a drone. Two machines with the same level can be unequally tolerable, and no acoustic method on this shelf addresses it.

The documents it sits next to

The directive on outdoor noise is the one that requires the figure to exist and to be marked. This document is how the number is obtained. Neither is the operator-position figure printed in a manual, which is a sound pressure level determined under yet another method.

Machine product standards — the pedestrian mower standard, the shredder standard, the hand-held machinery family — supply the operating conditions, and in some cases modify the measurement surface.

Which shelves it reaches here

It is cited by the machine families whose listings carry a sound power figure at all: the four mower shelves, the robots, the shredders and the blowers.

Where a review here reports a noise figure it names the quantity, and where a maker publishes only a bare number the page says the quantity is unstated. That is not pedantry on a shelf where sound pressure at the operator’s ear and guaranteed sound power routinely differ by well over ten decibels, and where both are printed as “dB”.