U.S. Marine Corps Level-Dependent Hearing Protector Assessment: Objective Measures of Hearing Protection Devices.

摘要

To characterize the effects of four level-dependent earplug-style hearing protection devices (HPDs), measurements were made of the passive attenuation of steady-state and impulsive noise, as well as the localization ability of listeners while they wore each HPD. Three HPDs were mechanical (Moldex BattlePlugs, 3M Combat Arms Earplugs, and the SureFire EP4 Sonic Defenders Plus). These triple-flange earplug-style HPDs provide level-dependent protection through the incorporation of a filter that allows the user to hear ambient sounds with minimal attenuation but protects against impulsive noises above about 105-dB peak. We tested the fourth HPD (Etym tic Electronic BlastPLG EB15) in the inactive condition only, but it can provide ambient environmental hearing via external microphones and uses electronic circuitry to limit or shut off the transmission of unsafe levels of noise. The three mechanical earplugs provided similar (8 14 dB) passive attenuation of steady-state pink noise presented at 105 dB A. One-third octave band analyses of this attenuation showed differences in the spectral profile at higher frequencies. The inactive EB15, tested with foam ACCU Fit ER38-14F ear-tips, provided 28 dB of passive attenuation. The devices all provided in excess of 30-dB attenuation of an impulsive noise presented at approximately 157-dB peak; in particular, the inactive EB15 provided more than 34 dB of passive attenuation. All four devices caused an increase in average localization error near 0 and 180 , most likely due to reversals. The average unsigned azimuth error did not differ significantly across the four earplugs, ranging from about 38 to 42. All four earplugs, as tested, provide sufficient protection against noise. The significance of the mechanical earplugs differential effects on the spectral content of lower-level steady-state noise on spatial orientation and, ultimately, user acceptance are discussed. Similarly, we note the limitations of inferences made from tests.