This second part is devoted to the identification of vortex sound sources in low-speed turbomachinery. As a starting point, the time-resolved evolution of the vortical motions associated to the wake shear layers (reported in the first part of the present study) is employed to obtain vorticity distributions in both blade-to-blade and traverse locations throughout the axial fan stage.
Following, the Powell analogy for generation of vortex sound is revisited to obtain the noise sources in the nearfield region of the fan. Both numerical and experimental databases presented previously are now post-processed to achieve a deep understanding of the aeroacoustic behavior of the vortical scales present in the flow.
A LES simulation at midspan, using a 2.5D scheme, allows an accurate description of the turn-out time of the shedding vortices, within high-density meshes in the blades and vanes passages, and a correct modeling of the dynamics of turbulence. Besides, thermal anemometry has been employed with a two-wire probe to measure the planar flow in the midspan sections of the fan. Statistical procedures and signal conditioning of velocity traces have confirmed experimentally the unsteady flow patterns devised in the numerical model.
The comparison of the rotor-stator and the stator-rotor configurations provides the influence of the wake mixing and the nucleation of turbulent spots in the distribution of the Powell source terms. Moreover, the relation between the turbomachine configuration and the generation of vortex sound can be established, including the impact of the operating conditions and the contributions of the interaction mechanisms.