Experimental and Computational Study of Vehicle Surface Contamination on a Generic Bluff Body

Abstract

This paper focuses on methods used to model vehicle surface contamination arising as a result of rear wake aerodynamics. Besides being unsightly, contamination, such as self-soiling from rear tyre spray, can degrade the performance of lighting, rear view cameras and obstruct visibility through windows. In order to accurately predict likely contamination patterns, it is necessary to consider the aerodynamics and multiphase spray processes together. This paper presents an experimental and numerical (CFD) investigation of the phenomenon.The experimental study investigates contamination with controlled conditions in a wind tunnel using a generic bluff body (the Windsor model.) Contamination is represented by a water spray located beneath the rear of the vehicle. The aim is to investigate the fundamentals of contamination in a case where both flow field and contamination patterns can be measured, and also to provide validation of modelling techniques in a case where flow and spray conditions are known.CFD results were obtained using both steady RANS and unsteady URANS solvers, combined with particle tracking methods. Steady RANS does not capture the wake structures accurately and this affects the contamination prediction. URANS is able to recover the large-scale wake unsteadiness seen in the experimental data, but the difference between the experimental and computational contamination distributions is still notable. The CFD is also able to provide further insight by showing the behaviour of particles of different sizes. Large particles are found to take on a ballistic trajectory and penetrate the wake. In contrast, small particles are shown to be less likely to become entrained into the wake.

Computational investigation into the sensitivity of a simplified vehicle wake to small base geometry changes

“…The Windsor model is known to produce a three-dimensional and highly unsteady wake meaning steady approaches such as RANS are insufficient at capturing the required timedependent turbulent structures to fully replicate the flow field [28]. The computational domain ( Figure 4) was defined to reflect the wind tunnel test conditions under which the validation data would be obtained.…”

A Parametric Study of Automotive Rear End Geometries on Rear Soiling

“…Whereas much published work [7,8,9,10] has focused on the final outcome with regard to the surface contamination, or on comparing CFD simulations with experimental work, the work presented here studies the variation of base contamination with changing roof slant angle. Roof slant angle variations were selected for the study because they are commonly employed as a drag reduction method [11].…”

“…The calibration is achieved by placing a calibration vessel similar to that used in [9,18] in the tunnel next to the model at the end of every test so that it is captured in the image. Figure 6 shows the variation of intensity with depth across the calibration vessel.…”

Surface contamination of cars: A review

“…Therefore, it is important to use higher-fidelity turbulence modelling than that provided by either a Reynolds-averaged Navier-Stokes (RANS) model or a URANS model. Hence, it is not surprising that Kabanovs et al 90 found that neither the RANS nor URANS turbulence models, combined with a Lagrangian particle model, can provide satisfactory rear-surface deposition patterns for a simple car-like bluff body. This was also seen to be a result of the poor base pressure distributions predicted by these methods, an important aspect of the physical problem, as Costelli 36 observed a correlation between contaminant deposition and regions of relatively high surface static pressure.…”