The effects of simplifications on isolated wheel aerodynamics

Abstract

a b s t r a c tIn order to study the aerodynamic forces and flow features of rotating wheels, compromises and simplifications are often made in wind tunnel testing, and more frequently so in numerical modelling. A CFD approach similar to that commonly used in industry was utilised to investigate common assumptions involving; the influence of geometric fidelity in wheel hub regions, ground representation, the modelling of the contact patch, and the effects of rotation on separation. It was found that the separation and wake characteristics were strongly influenced by the rotation of the wheel; the separation point changed by as much as 90% compared to a stationary wheel, and drag was close to 20% less -downforce was approximately 40% greater. In addition, the modelling of the contact patch, treated here as a small step to facilitate skew-free meshing necessary for a reliable converged result, was seen to cause up to a 52% difference in predicted lift characteristics, and an increase in the step of just 2 mm decreased the maximum wake thickness by close to 50% -considerable changes stemming from superficially-minor simplifications. Including indented wheel hubs proved to be more influential on the production of vortices and wake structures, causing the merging of previously-separate vortex structures. The results point to a need for very careful evaluation of the goals of any study when determining which simplifications can be made in both physical testing and numerical analysis.

Artificial intelligence-based Monte-Carlo numerical simulation of aerodynamics of tire grooves using computational fluid dynamics

“…Geometry used by Fackrell and Harvey was simplified for numerical analysis by Axon et al (1998). Diasinos et al report that in order to better study the impact of factors like groove dimensions and corner radii, etc., the simplification of model geometry is very useful (Diasinos et al, 2015). Therefore, based on the recommendations of Axon et al, a simplified tire geometry for current experimentation has been deployed as shown in Figure 1.…”

“…Both the rotating and the separating flow phenomena occur in our computer-simulated experiments. The comparison between different turbulence models on tire model is given by Diasinos et al (2015). The Realizable K-ε RANS model is used to solve complex flows by different studies (Doig and Barber, 2011;Blocken and Toparlar, 2015;Diasinos et al, 2017).…”

Computational modelling of a solid and deformed automotive rotating wheel in contact with the ground

“…There are four main computational techniques of comprehensively modelling the flow around the rotating wheels. The rotating boundary condition 3,10,17 and moving reference frame 18 are popular due to their simple implementation, mainly employed for steady state approximations. The sliding mesh approach 19 enables time-depended simulations where fixed and moving spatial domains can be considered.…”

“…Contact modelling would be ideally represented by two tangent surfaces; however, in practice, this usually leads to the incurrence of extremely small elements in the vicinity of the contact area, with element of high skewness as McManus and Zhang 10 and Mears and Dominy 15 experienced with their computational models. One of the most popular solutions is the use of an artificial step as a transition between the wheel and the ground; Diasinos et al 3 and Heyder-Bruckner 17 demonstrated this technique and its implementation. Axon et al 1 defined a wall velocity on the contact patch so that the discontinuities between the tangential velocity of the tyre and the moving ground are minimised.…”

“…Axon et al 1 defined a wall velocity on the contact patch so that the discontinuities between the tangential velocity of the tyre and the moving ground are minimised. Flow structures developed downstream of the wheel are directly affected by the step’s height as highlighted by Diasinos et al 3 Another approach applied by Sprot 22 and Knowles 11 consists on trimming the wheel to simulate the contact with the ground as a flat surface. The front and rear jetting phenomena are correctly predicted, although the strength of the front one is mispredicted compared to experimental data.…”