An Experimental Study of the Effects of the Test Configuration on the Noise Generated by Axial Flow Fans

The present paper concerns the effect of the test configuration on the noise generated by axial flow fans employed for automotive cooling systems. During the development of a fan, a number of noise tests are usually taken on different configurations, whose characteristics depend on the design step. Often, significant discrepancies are observed between the results of different tests, resulting in difficulties in transferring and generalizing the measured SPL level spectra. Such differences regard both the broad band noise and the tonal one. In the present paper the effects of the following aspects are considered: the rotor balancing method, the geometrical tolerances, the rotor mounting system on simplified configurations (with and without downstream stator) and the production configuration (a stationary plastic shroud) without heat exchangers. The balancing method affects the high frequency part of the SPL spectrum, due to the broad band noise, while the mounting system may affect the OASPL and the low frequency part of the SPL spectrum, due to the tonal noise. The geometrical tolerances do not affect the generated noise.