Computational investigation into the influence of yaw on the aerodynamics of an isolated wheel in free air

Abstract

This paper details a computational investigation into the influence of yaw angle on the near and far-field aerodynamics of an isolated wheel in free air. Unsteady Reynolds-averaged Navier-Stokes was used as the primary analysis tool to analyse the complex flow features around this configuration, with principal vortical positions and magnitudes, overall lift, drag, and side force coefficients, as well as on-surface pressures characterised within the work presented. Fundamentally, the flow was found to be highly vortical in nature, particularly within the wake region of the wheel, with changes in yaw shown to have a significant impact on the surface flow, wake structure, as well as calculated force coefficients.

Computational investigation into the influence of yaw on the aerodynamics of a rotating wheel in free air

“…However, as the flow opposes the direction of wheel rotation, the flow separates earlier (B) at θ = 264° as the intense up-wash (C) along the rear surface of the wheel (induced by wheel rotation) interacts with the upper shear layer (SLR) which propagates into the rear wake. This wake region was also found to be shifted upwards, more towards the top of the wheel, when compared to the baseline stationary wheel in free air [Kothalawala and Gatto (2015)]. The wake behind the stationary wheel is entrained within the shear layers formed after separation between θ=140° and θ=220°, whilst the wake behind the rotating wheel is entrained between the upper flow separation region at θ=264° and the lower flow detachment region at θ=145°.…”

“…Further downstream, the vortices were found to laterally translate away from each other, as V1R translates in the -x direction, whilst V2R translates in the +x direction. With this translation of the two vortices in the -x & +x direction for V1R & V2R respectively, both vortices also continue to translate upwards in the +y direction, as illustrated in Figure 7, representing a taller wake when compared to the stationary case [Kothalawala and Gatto (2015)], and is also expected on a rotating wheel in contact with the ground [Fackrell (1974), McManus and Zhang (2006), Mears (2004)]. From flow-field interrogation, vortex core vorticity magnitudes were also found to be similar between the two vortices comprising a difference of only 12% at z/d = 0.75 as d/U∞ = 5.05 for V1R and d/U∞ = 5.64 for V2R.…”

“…Where possible, this validation involves the use of both mean surface data as well as near-field wake data, however, particularly for the latter, limited data exists for comparison. To obtain results for the baseline stationary case, the methodology was identical to that outlined in the previous section albeit with a stationary, non-moving wall boundary condition for the wheel, as discussed further in Kothalawala & Gatto (2015). As an initial comparison, Figure 4a shows the mean co-efficient of surface pressures along the wheel centreline (x/d=0) obtained from the URANS study [Kothalawala & Gatto (2015)], plotted against centreline experimental data manually extracted from literature [Zhang et al (2013), Lazos (2002)].…”

“…To obtain results for the baseline stationary case, the methodology was identical to that outlined in the previous section albeit with a stationary, non-moving wall boundary condition for the wheel, as discussed further in Kothalawala & Gatto (2015). As an initial comparison, Figure 4a shows the mean co-efficient of surface pressures along the wheel centreline (x/d=0) obtained from the URANS study [Kothalawala & Gatto (2015)], plotted against centreline experimental data manually extracted from literature [Zhang et al (2013), Lazos (2002)]. As illustrated in Figure 4, there is excellent agreement to Zhang et al (2013) over most of the wheel centreline circumference with maximum deviations found to be typically less than CP0.05.…”

“…Moreover, the intense up-wash behind this rotating wheel was found to be greater in vertical velocity (Y/U∞) when compared to the up-wash and downwash present behind the stationary wheel which approximately provided Y/U∞ = ±0.31 respectively. However, considering the flow-field on either side of the horizontal centreline of the wheel, the observed rear wake illustrated only two dominant vortical structures, indicating that a fundamental characteristic of the application of rotation transforms the four vortex rear wake initially observed on a stationary wheel in free air [Kothalawala and Gatto (2015)], to a two vortex wake positioned on the upper rear wake. This is also a fundamental change when compared to the number of vortical structures observed on the rear wake of a stationary wheel in contact with the ground [Fackrell J.E., (1974), McManus, J. and Zhang, X., ( 2006)].…”