Search Results

The turbulence environment in the real world is known to be significantly different to that found in a typical automotive wind tunnel. Various studies have shown that raising the level of freestream turbulence has an effect on the forces on generic bluff bodies and real vehicles. Previous work at Loughborough has shown a significant effect of raised freestream turbulence on edge radius optimisation using measurements of forces and moments, and in this paper the underlying changes in the flowfield are investigated using PIV. Results are presented of the flowfield around the leading edge radius of the generic bluff body used in the previous work. The effect of changing the Reynolds number is investigated in the clean tunnel (0.2% turbulence), and it is found that, when the radius is small, there is a significant separation that persists up to a high speed, and then abruptly collapses.

Various studies have shown that the level of wind noise experienced inside cars on the road in unsteady conditions can be substantially different from that measured in wind tunnel tests conducted using a low turbulence facility. In this paper a simple geometric body representing the cabin of a passenger car has been used to investigate the effects of free stream turbulence, (FST), on the A-pillar vortex flowfield and the side glass pressure distribution. Beneath the A-pillar vortex, both mean and dynamic pressures are increased by FST. The unsteady pressure can be associated with wind noise and the flow visualization shows the peak unsteadiness is related to the separation of the secondary vortex.

This paper presents the findings from a numerical study of a gasoline direct injection engine flow using the Large Eddy Simulation (LES) modelling technique. The study is carried out over 30 successive engine cycles. The study illustrates how the more simple but robust Smagorinsky LES sub-grid scale turbulence model can be applied to a complex engine geometry with realistic engineering mesh size and computational expense whilst still meeting the filter width requirements to resolve the majority of large scale turbulent structures. Detailed description is provided here for the computational setup, including the initialisation strategy. The mesh is evaluated using a turbulence resolution parameter and shows the solution to generally resolve upwards of 80% of the turbulence kinetic energy.

It has been recognised that the ideal flow conditions that exist in the modern automotive wind tunnel do not accurately simulate the environment experienced by vehicles on the road. This paper investigates the effect of varying one flow parameter, freestream turbulence, and a single shape parameter, leading edge radius, on aerodynamic drag. The tests were carried out at model scale in the Loughborough University Wind Tunnel, using a very simple 2-box shape, and in the MIRA Full Scale Wind Tunnel using the MIRA squareback Reference Car. Turbulence intensities up to 5% were generated by grids and had a strong effect on transcritical Reynolds number and Reynolds sensitivity at both model scale and full scale. There was a good correlation between the results in both tunnels.

A drag coefficient, which is representative of the drag of a car undergoing a particular drive cycle, known as the cycle-averaged-drag coefficient, has been previously developed. It was derived for different drive cycles using mean values for the natural wind. It assumed terrain dependent wind velocities based on the Weibull function, equi-probable wind direction and shear effects. It did not, however, include any effects of turbulence in the natural wind. Some recent research using active vanes in the wind tunnel to generate turbulence has suggested that the effect on drag can be evaluated from the quasi steady wind inputs. On this basis a simple quasi-steady theory for the effect of turbulence on car drag is developed and applied to predicting the cycle-averaged-drag coefficient for a range of cars of different types. The drag is always increased by the turbulence but in all cases is relatively small.

Understanding and predicting in-cylinder flow structures that occur within compression-ignition engines is vital if further optimisation of combustion systems is to be achieved. To enable this prediction, fully validated computational models of the complex turbulent flow-fields generated during the intake and compression process are needed. However, generating, analysing and interpreting experimental data to achieve this validation remains a complex challenge due to the variability that occurs from cycle to cycle. The flow-velocity data gathered in this study, obtained from a single-cylinder CI engine with optical access using high-speed PIV, demonstrates that significantly different structures are generated over different cycles, resulting in the mean flow failing to adequately reflect the typical flow produced in-cylinder.