It’s Clear to ‘C’

Why you should not use the SNR number for hearing protection assessments.

The Control of Noise at Work Regulations 2005 require you to assess the effectiveness of hearing protection if you are using it to help control exposure of employees to noise in the workplace. There are three ways to do this, so which one should you use and why should you avoid the SNR number!

First of all, the HSE guidance to the regulations states, ‘The information (SNR) is intended as a guide rather than a substitute for using one of the methods in Appendix 1 (HML or Octave Band), and in particular will not be appropriate if there are significant low-frequency components to the noise in question.’ There is a further justification and a very practical reason why this isn’t even onerous to do.

There are three methods of assessing hearing protectors; The Single Number Rating (SNR), which uses a single number measure of hearing protector performance and compares it to the L_Ceq noise level (note this is ‘C’ weighted!), the High, Medium, Low method (HML), which takes the frequency of noise into account and uses the L_Ceq and the L_Aeq measures of noise. Finally, the Octave Band method, which is a more detailed analysis and will return the most accurate answer based on comparing the noise level in eight octave bands with the performance of a hearing protector at those frequencies. The detail of these methods is explained in the regulation guidance document from the HSE, L108.

The table below shows the data for a Hellberg Secure 2 headband earmuff, which shows all the frequency data in the top 4 rows and the HML and SNR data in the last row. The Mean Attenuation is the result of testing the device and the Standard Deviation (Std. dev.) is the statistical correction for the way the testing is carried out. The Assumed Protection Value (APV) is simply the Mean minus the standard deviation.

Frequency Hz	63	125	250	500	1000	2000	4000	8000
Mean Attenuation	18,2	13,6	21,8	30,7	39,4	35,8	37,6	40,0
Std. dev.	5,4	3,4	2,7	3,1	3,0	2,9	2,8	4,8
APV	12,8	10,2	19,1	27,6	36,4	32,9	34,8	35,2

HML/SNR

35dB

28dB

18dB

SNR:

30dB

Secure 2H Headband; EN 352-1:2002, Weight 248g

Ear defenders virtually always perform better at high frequencies compared with low, so clearly, if you can account for the make-up of the noise, then you will get a better assessment of the effectiveness of a given ear plug or muff, especially when you consider that much industrial noise has considerable low-frequency elements.

The Single Number Rating of a hearing protector tells you how much that device will reduce the noise level by on average in a ‘normal’ noise environment with neither any dominant low nor high frequencies. This is useful for comparing one protector with another but is limited when assessing the actual performance. As we are dealing primarily with industrial noise, then the ‘C’ weighted measurement is used, although this does not fully compensate for the real-world. The assessment is simply the ‘C’ weighted Leq minus the SNR number with an extra 4dB added in the UK (as recommended by the HSE) to compensate for real-world fitting errors.

The HML method, which stands for High, Medium, and Low, where the hearing protector is given a rating based on its performance in those frequency ranges (you can see this in the table earlier). This method requires you to have measured the ‘A’ weighted as well as the ‘C’ weighted levels, but as we already know from the SNR method, you need the LCeq for that, and you will always need the LAeq for your exposure assessments anyway. It is quite easy to argue, then that as you need both these numbers anyway and the HML assessment is better, then there is no reason why you should ever need to do an SNR hearing protector assessment. There is an example of an HML result from www.TheCastleCloud.com below.

If you own a sound meter with the capability to measure octave bands, then you can use the full frequency data for the APV, which will give the best result in terms of accounting for the frequency of the noise in a particular environment. In essence, each APV value is subtracted from the octave band and then the resulting numbers are combined back together to give an overall protected sound level (see graph below). As with all the methods, in the UK this is further de-rated by adding 4dB for real-world fitting errors.

From the above, LAeq = 85.79 while APV = 75.71

There are plenty of software packages available that will carry out these calculations for you and the Castle dBAir sound meter even has a database of hearing protectors built in to make the assessment available on-the-spot. You simply measure the sound, select a make and model of ear defender and the answer pops up on-screen.

In conclusion, the minimum requirement for hearing protection assessment requires a measurement of the ‘C’ weighted sound level, which combined with the ‘A’ weighted level from your exposure assessments means you will always have the data you need to use the HML method. If you have octave bands measurement capability, then you should certainly use it! The SNR number is useful for comparing one hearing protector with another, but it should really be reserved for that purpose!