TLDR
There are different kinds of sound levels which isn't obvious from just quoting a decibel value. if you are talking about sound pressure level you must also give a distance. It's also important how the sound pressure level was measured. 80 dB by itself doesn't mean a lot.
What is sound?
Most text books on acoustics invariably begin with a chapter called "What is sound?". Well, maybe not those exact words - sometimes they use more fancy works like: "Fundamentals of Vibration" which starts off Kinsler's "Fundamentals of Acoustics" or "The nature and behaviour of sound" which Bob Peters begins his textbook "Acoustics and Noise Control". But to save you the reading, it basically boils downs to
things vibrate
those vibration travel through a medium (air) and
our inner ears convert those vibrations into small electrical signals that our brains 'hear'
*The above animation is wrong by the way. Probably deliberately so because it's easier to visualise waves going up and down, but really they go back and forth.
This is what some text books have to say about it (use the left and right arrow keys to see more text):
What is Noise?
Noise and sound often get used interchangeably but they aren't the same thing. The one sentence definition is: noise is sound we don't want to hear.
For some reason, I think acousticians are fixated on illustrating noise with lawnmowers. Or maybe it's just me - but the image of a lawnmower in an old B&K publication always stuck in my head.
What isn't a "dB"?
It gets a bit more complicated when trying to define things in terms of a quantity. The problem is that we feel we know what sound and noise is - mainly because it is part of their everyday lives - we then try to treat decibels (abbreviated dB) in the same way as other things like cm or kg or mph. And this is where people trip up. Decibels are not a measure of anything. It's not a quantity like pints are a quantity of milk*.
* at least in the UK where we use a confusing mix of SI units and imperial units
What is a "dB"?
The 'd' stands for 'deci' as in one tenth of something and the 'B' stands for 'Bel' and they get smashed together as 'decibel' or dB. Bel is named in honour of Alexandra Graham Bell - hence the B is capitalised as a proper noun. So there are ten dBs to a B or to say it the other way round: one dB is a tenth of a B.
A 'Bel' (never used without the deci), is the common logarithm of the ratio of two numbers. That's one number divided by another number and then log whatever the result is. You would then times that by ten to get a decibel. You can do it on your calculator at home like this -> -> ->
The reason we use logarithms is because they compress a very wide range of ratios into something a little more manageable. Logarithms changes multipliers into additions. Add 10 dB to a something and it is the same as multiplying by 10. An increase in 20 dB is an increase by 100 time and an increase by 30 dB is an increase by 1000 times. So 80 dB is a ratio of 1 to 800 million!
And in all that explanation I have not made one mention of sound or noise. That's because decibels are not exclusive to sound. They can be used as a measure of many things - electronics, optics, video and imaging, to name a few. All of which need a wide range of quantities compressed into a range of numbers that are a little more manageable than using millions and billions.
So how does sound pressure come into it?
Firstly, the actual units for sound pressure are 'pascals' - the same unit we use for any normal sort of pressure - like water pressure or atmospheric pressure. But, for sound pressure we normally mean the root-mean-square (rms) of the pressure. This is because sound is where the pressure is constantly changing - its the changing in pressure that we are hearing - see the picture to the left.
The sound pressure level (note the extra word 'level' which is what dBs are used for) is the rms sound pressure divided by 0.00002 pascals, then squared, and then you do the 10 x log thing on it. That's a bit of a mouthful. It's easier to write as an equation:
Where Lp is the sound pressure level;
Prms is the rms sound pressure; and
Pref is 0.00002 pascals
That's quite technical for an easy read blog so you can just forget the equation, But, an easy way to put it is that the sound pressure level in decibel describes the sound you hear relative to the quietest thing we can hear. So 80 dB is 800 million times more than the quietest thing we can hear.*
And yes, you can technically get negative decibels - and that's not sound being sucked away - it's just sound pressures less than 0.00002 pascals.
* This is not quite correct but close enough to get the gist of it
Obviously we mean sound pressure!
If we are talking about decibels in the context of sound levels then isn't it obvious we mean it as the ratio of sound pressures and not whatever they use for optics? But hang on. It's still not as simple as that!
We also use decibels to describe sound power levels which is measured in watts and not pascals!* Sound power is how much power is in the sound being generated. You might be familiar with audio amplifiers delivering power in watts**.
But even if we were sure we were talking about sound pressures, we still need to know how far away you are. Think about it. A lawnmower a mile away can't be heard. But if you put your ear right up to the motor, it might be a lot more than 80 dB. So if we are talking about sound pressures we must also know the distance at which it was measured. E.g. 80 dB at 10 m. If we are talking about sound powers then there is no distance involved because we are talking about a power. The fact that we use decibels to describe both caused all sorts of confusion.
But even if we do say 80 dB at 10 m, that's still not the end of the story! Is that the instantaneous sound level? Is it the maximum sound level? Is it the average over some time period? Each of those has a distinct unit which in itself is not straightforward to define. And we use decibels for all of them! Usually we use some other letters to tell us exactly what we are referring to - such as LAeq - which is sort of the average - also not a straightforward definition.
*There is also a much less frequently referred to 'sound intensity level' which is the flow of sound power in a direction.
**Technically this is not a measure of sound power because there are losses as the signal gets converted to sound.
I measured the sound pressure level with my phone
Or maybe you bought a cheap sound level meter from Amazon. Is it good enough? I think it's more than likely that your phone app or cheap sound level meter is getting all the calculations correct... maybe. But the main issue is likely the microphone and the placement of the microphone in the phone. The sensitivity of the microphones is probably unknown to the app and very unlikely to be consistent across time and frequencies. Also, how you are holding the phone? What hard surfaces are near by? What reflections are coming off you? These are all things that are going to come into play.
Maybe you'll get a decent measurement. But, more often than not the acoustic consultant will nod politely as you tell them about it and then they'll do their own measurements.
The specification sheet for my lawnmower has the sound level
We tend to take what's written in specification sheets more seriously. After all, if they have reported a number on a document with the equipment and it's wrong then surely that's on them. Well maybe. I don't what the penalty for getting it wrong is. I've never heard of a case where it was discovered the reported sound level was wrong which led to some kind of punishment.
But even so, we often run into all the same definition problems. Sometimes it's not stated whether it's a sound power level or a sound pressure level. Sometimes the sound pressure level doesn't give a distance. More often than not it's in there for noise at work reasons and it is given at the operators position - which is no help if that's inside a cab or we don't even know where in relation to the machinery. The operating conditions are almost never defined.
But at least it was done by someone suitably qualified following a standardised methodology - right? Well sometimes. Sometimes when you dig a little, no one knows who did the measurements or how or when they were done. Then you ask yourself was it some gap year student standing next to the lawnmower with his phone?
Specification sheets are hit and miss. There are times I have had to resort to using sound level measurements of something similar that I know is reliable and disregard the specification sheet altogether.
I want to know more?
Sound is equally fascinating and confusing if you are not used to the concepts. You don't need to be an expert to understand it but it's not a topic that gets a lot of exposure. So it's not very well understood by most people.
Our cars have speedometers to tell us how fast we are going, we have kitchen scales to measure weights, jugs to measure volume, and we know how to use a ruler to measure the length of something. Why shouldn't you be able to point you phone at your lawnmower to get an accurate measurement? And when 10 kg + 10 Kg = 20 kg, why shouldn't 10 dB + 10 dB = 20 dB (it actually equals 13 dB). Well there you go. It feels like it should follow all the normal rules but it doesn't
If you want to find out more then one of these books is a good place to start (click on the picture for a link to where you can buy them). If you want guidance on measuring sound then this publications provides some best practice. You also can't go wrong with one of these courses.
Commentaires