The modelling of rotating parts, such as axial fans, is one of the main challenges of current CFD simulations of industrial applications. Different methods are available, but the most commonly used is the multiple reference frame (MRF) method. This paper investigates how different flow properties, such as temperature, pressure and velocity, develop when passing through the MRF domain. The results are compared to the more physical rigid body motion (RBM) approach. It is found that the MRF method transports the upstream properties with the streamlines of the relative velocity from the upstream to the downstream interface. This leads to a non-physical rotation by an angle that is dependent on the length of the domain and the ratio between axial and tangential velocity in the MRF region. The temperature field is more affected than the flow field, since wake structures from upstream obstacles are destroyed due to the wake of the blades. Downstream structures affect the flow in the upstream region by an increase in static pressure, which causes the streamlines in the MRF zone to slow down. Depending on the size of the obstacle, this can cause substantial distortions in the upstream and downstream flow field.

No abstract

In this paper, a simplified underhood environment is proposed to investigate the air flow distribution in a vehicle-like set-up and provide high quality measurement data that can be used for the validation of Computational Fluid Dynamic methods. The rig can be equipped with two types of front openings representative for electrified vehicles. Furthermore, it is possible to install differently shaped blockages downstream of the fan to imitate large underhood components. The distance between the blockages and the fan can be varied in longitudinal and lateral direction. The measurements are performed with Laser Doppler Anemometry at a fixed distance downstream of the fan. The results show that the lack of an upper grille opening in the configuration for a battery electric vehicle has a notable impact on the flow field in the reference case without any downstream blockage. However, the differences in the flow field between the two front designs become less when a downstream obstruction is present. The longitudinal and lateral position of the blockages have a minor impact on the flow field compared to the shape of the obstacle itself.

No abstract