Ljudexponering vid användning av hörselskydd med intern kommunikation

Sammanfattning: The Swedish armed forces, Försvarsmakten (FM), is the governing body in charge of defending the sovereignty of Sweden. In FM, a large part of the personnel is exposed to high levels of noise as part of their work with machines and weaponry. The Swedish authority on workplace related hazards, Arbetsmiljöverket, has set out limitations for noise exposure during work in the publication: AFS 2005:16. To reduce the total noise exposure of the employees, FM have implemented multiple changes in their work. These changes include the use of hearing protection devices (HPD) but also include changes in routine and equipment. Multiple versions of the hearing protectors include radio communications to facilitate communication while maintaining a high level of noise reduction. The methods set out in AFS 2005:16 to measure noise exposure are not applicable when sound is generated inside the ear canal. Therefore, the total noise exposure cannot be determined accurately with regards to the set limitations. The thesis aims to determine a method to measure the total noise exposure in representative conditions and how the use of this method along with other objective measures of sound to set system settings. The method of choice is to use a HATS or MIRE measurement in accordance with ISO11904- 1:2002 or ISO11904-2:2004 to properly capture the influence of the operator and calculate the contribution from the communication systems. The contribution from external sources need to be evaluated separately as the attenuation of hearing protection devices may not be properly measured when using MIRE or HATS. When such contribution is calculated, insertion loss (IL) determined using the REAT method should be used. The exposure using three weapon systems was measured using a HATS at representative head position and a selection of HPD. The resulting IL was compared to other methods of measuring IL and total exposure using these results. The IL in in-situ measurements was influenced by the presence of a disturbance in the HATS and consequently overestimated exposure. IL was also measured in a reverberation chamber and an echo-free chamber using a HATS. The resulting IL seems to differ from REAT measurements and did in some cases severely overestimate the IL. The thesis also simulated the influence of three idealized types of noise on the Speech Transmission Index (STI), an objective measure of speech intelligibility. To measure STI using HATS or MIRE measurements, a transfer function like the one from ISO11904-2:2004 needs to be applied if noise or HPD:s will be used. Previous work recommends a speech-to-noise ratio of at least 12dB to achieve good speech intelligibility. The rule of thumb seems to have some advantages when using a simulated Matrix Transfer Function but does overestimate the required level for low noise levels. For higher levels, excellent STI (0,75) does not seem to be achievable. The required signal-to- noise is dependent on both the spectral content of the noise as well as the level, which both is influenced by the use of hearing protection devices. To select a proper signal to noise level, a representative recording of speech using relevant equipment is required. The influence of noise can be added separately using the methods in IEC60268-16 and REAT measured attenuation.