Having a safe and comfortable work environment is not only prescribed in law, it makes good business sense.
NZ sound laws are based on industry Standards.
Until 2007, IEC-651 (renamed to IEC-60651) defined standards for sound level meters. The current specification is IEC-61672-1. For meaningful, repeatable sound measurements, it’s a good idea to look at meters manufactured to comply to the latest Standard.
The 2013 IEC-61672 Standard is divided into 3 sections:
gives electro-acoustical performance specifications for three kinds of sound measuring instruments:
- Time-weighting sound level meters that measure exponential-time-weighted, frequency-weighted sound levels
- Integrating-averaging sound level meters that measure time-averaged, frequency-weighted sound levels
- Integrating sound level meters that measure frequency-weighted sound exposure levels.
provides details of the tests necessary to verify conformance to all mandatory specifications for the above types of meters
describes procedures for periodic testing of the above types of meters.
Measuring noise levels is more difficult than it might appear at first. Human hearing is far from linear; we hear much more poorly at low frequencies (bass) and very high frequencies (treble). Beside frequency, human hearing is also non-linear in respect of perceived volume. Human hearing is in fact logarithmic such that a doubling in Sound Pressure Level (SPL) is perceived by us as an only moderate increase in volume.
The Decibel is a logarithmic unit of measure used to compensate for this effect. Because the scale is logarithmic, it’s worth remembering that an increase of a few decibels (dB) represents a very large change in perceived volume.
As mentioned earlier, human ears are non-linear with respect to frequency (pitch) as well as volume.
As early as the 1920’s a pair of scientists developed the Fletcher-Munson curves to show human hearing perception at different frequencies and volumes.
What the chart shows it that at very quiet volumes, one cannot hear low bass, but at medium high frequencies sound can be easily heard. (Some evolutionist theory suggests that human hearing is keenest in the range of spoken word being 1 to 3 kHz)
As the volume level increases, this effect reduces. This is why many sound systems (especially automotive systems) have a button marked “Loudness”. This boosts low bass and high treble to compensate for low volumes, but is commonly used incorrectly to cause a preponderance of bass.
Hence using a linear noise measurement instrument, although accurate, gives a very poor representation of the experience of a human listener.
Sound level meters are used to check sound levels. If we were to use a sound level meter with a linear or flat response, this would accurately show the sound level in the room. This is useful for some tasks, but if we’re wanting to look at what human beings experience, it is not. Remember our hearing is NOT
linear or flat.
Hence we add ‘weighting’. We modify the frequency response of our instrument to mimic our human perception so we can measure the effect of what a person will experience.
A-weighting mimics human perception. The Fletcher-Munson curves show that we hear more poorly at low frequencies, thus we make the instrument less sensitive at lower frequencies. (Consider the A-weighting curve, blue, in the graph. Note how the sensitivity to sound rolls away from 1000 Hz and down)
Nearly all sound level meters will have switching for “A” or “C” weighting. If we want to measure the sound level as a human will perceive it, we’d use A-weighting. If we want to measure the actual sound pressure level in an absolute sense, we use C-weighting which is virtually linear.
Safety standards from Worksafe NZ or other regulatory bodies will define whether one is to use “A” or “C” weighting.
Incidentally the blue whale is not only the largest, but also the loudest animal in the world. They generate underwater sound levels of up to 180 dB! Which explains why they can hear each other half way around the world. Follow this link
for an interesting video of whale song.
A detailed analysis of the standards relating to noise in the workplace is beyond the scope of this article. Good documents may be found on the Worksafe NZ website by following this link
But to summarise and paraphrase, 85 dB (A) is the point at which exposure needs to be limited and protection used. It’s surprising how easily this level can be reached! Close exposure to a vacuum cleaner for example.
Incidentally the term ‘close exposure’ is apt. Distance-to-source plays a large role in the sound level. Sound from a point source follows the ‘inverse-square law’, such that every doubling of distance, halves the volume.
So when reading specifications, take careful note of distances quoted.
Calibration of Sound Meters
Many users do not require traceable calibration of these instruments, however it is good practice to check the instrument prior to use. Portable sound level calibrators are small sound generation devices that produce a fixed tone and sound level, usually at 2 levels, to check the performance of your sound level meter. If a higher level of calibration and traceability is required, please refer to www.ianz.govt.nz for a list of qualified laboratories.
‘Point source’ is mentioned above, as you may have noticed line-array speaker’s at large concerts, or column speakers in churches or even electrostatic speakers in top home hi-fi systems. These three examples are line-sources as opposed to point sources and don’t obey the inverse square law; decaying at a far slower, linear rate (hence their use in these applications).
Because of the logarithmic nature of human hearing, audio amplifier power out specifications become interesting. For example if one has an amplifier rated at 10 watts RMS output, how many watts would one require to double the perceived volume? One might assume 20 watts, but that’s incorrect. A doubling of amplifier power produces a very mild 3 dB increase in the volume we hear. 100 watts (i.e. 10 times!) is the power required to achieve twice the volume. (3 dB is sometimes described as the smallest easily perceptible change in volume. 1 dB is the smallest possible to perceive change in volume).
The relationship is defined by the formula 10 X log (Power 2/Power 1).
Some years ago a loudspeaker company (Cerwin Vega) did very well in certain markets by making speakers that were much more efficient. In this way they were able to achieve far louder volumes than most other speakers, regardless of their power output.
Typical speakers have an efficiency of around 80 to 90 dB for an input of 1 watt RMS. Cerwin Vega made speakers with an efficiency of 100 to 102 dB for an input of 1 watt RMS so that it was a doubling of volume using the same amplifier!
In the 80’s a classic retail demonstration was to play a Sony Walkman direct into a set of Cerwin Vegas!