Search

The aim of the work presented in this paper is to test, in the field, a method for the prediction of structure-borne sound emission from resiliently mounted machines. The vibrational power injected into the machine seating is the structure-borne equivalent of the (airborne) sound power. Initially, an error analysis is carried out to show the sensitivity of the measurement method to various sources of error. The vibrational power, which is assumed to have a vertical component only, through the mounts of two compressor installations is then measuredin situ.The prediction requires knowledge of the dynamic properties of the resilient mounts and the machine seating (floor), together with the free velocity of the machine. These quantities are measuredin situfor the two compressors. Good agreement between the measured and predicted power is obtained in the first case, particularly at frequencies below 250 Hz. In the second case, discrepancies are attributed to transmission paths not accounted for in the prediction such as flanking transmission via pipes. This raises the possibility of the measurement and prediction method being used together to demonstrate the presence of a significant flanking path. Power from rotational motion is shown not to contribute significantly in the frequency range up to 200 Hz.

The benefit to an individual worker of a noise control measure is assessed by comparing their personal daily noise exposure before and after modifications. However, in complicated situations, such as factories, several workers' exposure may be reduced by the same noise control measure to differing degrees, and the overall improvement cannot be quantified so easily. In the paper a procedure is proposed in which the benefit of a noise control measure to all workers is given as a single figure rating. Proper before and after comparison of the overall improvement is then possible. This rating can also be used to quantify the cost effectiveness of various options. An example calculation is given.