Sweep-angle effect on low-order acoustic prediction for low-speed fans

Abstract

The low-speed fans that are used for automotive engine cooling contribute to a significant part of the global noise emitted by the vehicle. Automotive engine cooling manufacturers are therefore interested in developing low-order noise prediction methodologies that include the radiated sound in pre-design optimization cycles. Two among the most influential broadband noise mechanisms characterizing the automotive axial cooling-fans spectra in operating conditions are the turbulence-impingement noise and the trailing-edge noise. We modelled them through the application of a semi-analytical method called Amiet’s airfoil theory, appropriately adapted via a strip-theory approach, in order to take the rotating motion into account. In modern fan designs, the fan blades are strongly forward-swept, meaning that the blade edges are not always perpendicular to the tangential-to-rotation velocity vector. This is particularly true at the tip of the blades where the sweep angle often reaches its maximum value, and where most of the noise is expected. Sweep was shown to reduce the noise emitted by isolated airfoils, but its effect on rotating machines is not yet well understood. Its prediction requires an extension to the classical Amiet’s theory, which has been implemented in this work, permitting to assess the importance of varying sweep on the acoustic far-field radiation. Amiet’s theory is therefore applied for the two noise mechanisms mentioned above, starting from steady CFD simulations, reproducing the ventilator geometry operating at its maximum-efficiency condition. In contrast with the classical unswept formulation, the predicted results considering the sweep variation along the blade span better compare with the experimental measured ones in the VKI anechoic chamber, especially for high frequencies.