Which is best, hand-held or wearable noise meters?

Noise meters generally come in two forms, hand-held and wearable (called dosemeters). Here is a clear unambiguous statement on how the two fit together, and the rest of the text afterwards is all waffle explaining this:

Hand-held trumps wearable. Wherever possible use a hand-held meter.

Wearable dosemeters are mainly for situations where you physically cannot use a hand-held noise meter.

That's as clear as I can be - hand-held meters are better than wearable dosemeters every time and there are many good reasons for this.

Wearable dosemeters are beloved of noise assessment consultants though as you can stick a load of meters on people and disappear to the pub for a nice long leisurely lunch, then appear a couple of hours later and have a set of lovely graphs to put in the report.

The idea that wearable meters are best and are needed for calculating exposures is one of the most commonly misunderstood elements of a noise assessment - they are useful, certainly, but very much play second-fiddle to a good hand-held meter.

Reasons why hand-held is best 

Interference - deliberate or accidental

If you are doing a noise assessment and pop a wearable meter like a doseBadge on someone and send them on their merry way, the second they walk away you have lost control of that assessment. You have no idea if their mates are shouting down it (they inevitably will be) or whether it has moved slightly and their collar is now rubbing against the microphone which will cause a massive spike in the results.

Even something as simple as the person wearing the meter talking a lot will impact the result as the microphone is close to their face - if you aren't there to see the data being gathered then for all you know most of what you've measured is an hour of someone chatting.

Identifying noise sources

A noise assessment is not just about measuring a number of decibels but also seeing what can be done to control noise and they have wandered off wearing a meter then you don't know what caused the result in the first place so identifying potential controls is impossible.

You also have no idea what job they were doing. They may supposed to be on a particular machine but you don't know if they've had a breakdown (the machine, not them personally), gone off to help a mate out, moved to another machine for a bit, or whatever. 

Basically, once they walk away all you get back is the exposure pattern with little confidence in how it was obtained.

With a hand-held meter you know what caused the noises you are measuring and know that there was no interference, deliberate or accidental.

My noise meters - one hand-held Pulsar Nova meter and ten wearable doseBadge dosemeters

My noise meters - one hand-held Pulsar Nova meter and ten wearable doseBadge dosemeters

One of the esteemed and rarified beings who teaches Institute of Acoustics courses once told me that personal wearable meters should never be used for less than three weeks for any one measurement. Within that, you do one week of measurements and then throw those results away as the individuals concerned are certain to have fiddled with the meter or their mates will have pratted about with it, usually  by shouting at it.

Once those results are binned then use it for another two weeks to iron out any other errors from their work routine.

He is rather extreme in his dislike of personal dosemeters but the point stands, they are very unreliable.

What about the whole 8-house average thing - don’t you need to wear meters for a long time for that?

Nope. You can easily get the figures you need to calculate a daily exposure from a hand-held meter, or from a dosemeter worn for a short time. For example, if someone does the same job all day and you measure the noise on it, after a few minutes the meter (worn or hand-held) will have settled to a steady figure. From that point on the value won’t change so it doesn’t matter if you measure for another seven hours or not.

Often jobs are cyclical and with a hand-held you just measure each part of the cycle and them combine it into a daily result, or with a wearable you measure for a few cycles, for example for an hour, and then use that to calculate a daily exposure.

If jobs are not running at the time you can easily measure all the parts separately, using sample runs if necessary, then manually calculate an eight hour exposure from that.

There are very very few cases where the only way to get a daily average exposure is to measure for a full shift. Although when I say ‘very very few’ it is that few I’ve never come across one.