Pure-tone, third-octave, and octave-band attenuation of ear protectors

Experimental confirmation is shown of the practical identity of measuring ear-protector attenuation for pure tones and for 1/3-octave bands centered at the same pure-tone frequencies. In typical broad-band industrial noise, variations of spectrum shape have a negligible influence on an ear protector's octave-band attenuation. The use of pure-tone attenuation values at octave-band center frequencies as estimates of octave-band attenuation can lead to errors exceeding 10 dB. In the octaves centered at 500 Hz and above, ear-protector attenuation should be measured at 1/3-octave center frequencies.