SNR method for choosing hearing protection
Or as it's best described, the 'just about good enough and at least it's something' method. If you are doing your noise measurements yourself I'd always recommend aiming to at least do the HML route, and if you are getting someone in and paying good money for the noise assessment I'd insist on getting the octave band route. But, assuming you can't do the HML one, or simply don't want to, the SNR is always there as a fall-back and it is considered 'suitable' and compliant with the Noise Regs so you won't be sharing a cell with Big Dave the Troubled Sex Pest for an insufficient assessment if you use it over the other two methods.
Despite how it sounds, SNR is not a foreign railway but stands for Single Number Rating, and guess what, it's a single number used for rating hearing protection and is usually found on all packaging for hearing protection, be it muffs or plugs.
(Here’s a good rule of thumb - if you are looking at buying some minor-brand hearing protector and they talk about SNR as ‘signal to noise ratio’, don’t buy it. If they can’t get the basics like this right then they don’t deserve your money. Yes, SNR can mean signal to noise ratio, but not in this context).
NRR vs SNR
By the way, just to confuse you, sometimes you may see the number, especially online, as NRR which is noise reduction rating. It's used the same way but the numbers are not interchangeable - NRR cannot be used as SNR. NRR is used by our Yankeeland brethren while SNR is the European one and they are calculated differently - differing weighting is given to the frequencies used in their calculation meaning NRR is not the same as SNR and the two are different.
Some hearing protection sold internationally may show one or both and the numbers for each may differ slightly occasionally. Just ignore NRR if you see it.
As an example, despite the spelling of the product name meaning that the marketing person who came up with it is in need of a good thrashing for this abomination against the English language, the Laser Lite plug is a very common one in use across industry and is a good choice. It's the one which has streaks in it and is vaguely like an old pear drop. Except don't suck it. Especially after it's been in an ear. This plug has an SNR of 35dB but an NRR of 32dB.
Using SNR to calculate protector effectiveness
For SNR to be of use, what you need is a single measurement of your noise, the C weighted sound level. To be clear, that is the average dB(C) value on your meter and NOT the average dB(A) or the peak dB(C) levels, but the C-weighted average reading which will be lower than the peak but probably just a bit higher than the dB(A) average. Using the C-peak by mistake may cause headaches as it is always higher than the average and will cause you to decide hearing protection is not good enough when it really is.
The calculation is then easy. Simply subtract the protector's SNR from the dB(C) noise level and Robert's your mum's brother, you've done it.
So if the protector has an SNR or 35 and your noise level is 104 dB(C), just do 104 - 35 and you get 69dB. This is what you would expect someone wearing the protector to actually hear. In reality they often don't wear it quite properly so it's good practice to add 4dB to the result to make an allowance for that, so in this case giving a result of 73dB. A perfect result for a hearing protector.
Alternatively, armed with your SNR and dB(C) numbers you can go to Noisemeters.co.uk Resources section and use their SNR calculator. Put your noise reading in one box and the SNR from the protector in the other and it will tell you if your protector is good enough.
It bears repeating as it's an easy mistake to fall into, but don't use dB(A) measurements for this, don't use the peak dB(C) and don’t use NRR numbers - you want the average dB(C) only so need to have a noise meter which will measure that.