According to the HSE website all 3 methods of Hearing Protection Calculator'calculation (HML, Octave Band & SNR) are reputed to be nearly similar in their accuracy in most situations. The only one that I have found consistently well off the mark is the 'SNR' tab calculator as it requests that the C-weighted noise level be used. This is also stated on the 'Introduction' tab. However, the "guide" at the bottom of the latter uses the A-weighted value which in fact brings most results considerably closer to the HML and Octave Band results for any of the examples I have used.
Are the website instructions mistaken where twice the dB(C) is requested but the example uses the dB(A)?

Thanks,

Peter

p.s. I have send my query to the HSE via mail this date and only saw that it could be 10 days before I see a reply.

Hi Phil,

Thanks for your comment. I have since found something an acoustic professional said last year - that using SNR is ok for any values up to 95 dB(A) and that Octave Bands should be used for anything above this - which seems to indicate that dB(A) is used and not dB(C). However, I will see what comes at me in the next few days.

Peter

As promised, I said I would post the HSE reply - which makes complete sense to me now.

"The spreadsheet gives three different methods of estimating the protection afforded by hearing protection - octave band method, HML method and SNR method. It is not necessary to use all three methods - any one will do, so long as one has the correct information to hand. All three methods give similar results within certain bounds (i.e. so long as the noise is not an extreme of low-frequency, for example). In many situations people do not have enough information to use any of the three methods. For example, only having the A-weighted sound level (e.g. LAeq). For these cases, HSE has devised a simple check/first approximation method based on the SNR method, but allowing use of the A-weighted noise level (the proper use of the SNR method requires having a C-weighted sound level, LCeq). This is what is outlined on the 'Introduction' tab. This 'simple check' method does not assess the protected level - it simply states what range of protectors is likely to be suitable. The reason you are finding apparent errors is that you do not have an LCeq value, it appears you have a C-weighted peak level. Therefore, in using both the HML method and the SNR method you are using the wrong type of input, and hence finding a difference between this and other methods.. From the data you have supplied it does not appear you have sufficient information to use other than the octave band method, or the simple check method. Remember, when using the octave band method, to use linear and not A-weighted octave band levels.