Interference of supports used for ground vehicle wind tunnel testing
In order to provide a correct aerodynamic simulation of a vehicle travelling along the ground, models are tested using rotating wheels in a wind tunnel with a moving ground. In the most common of moving ground configurations the model is supported by an overhead strut, usually designed as an aerofoil profile to minimise interference, with the wheels supported by lateral stings hinged to mounts outside the span of the moving ground plane. ๒ using this type of configuration it is assumed that the presence of the intruding supports do not markedly affect the aerodynamic behaviour of the model, but this assumption is not always valid. In order to quantify interference effects from model supports, a range of models were tested over a stationary ground plane mounted to an under floor balance. Each model was tested with and without mock struts and stings, which do not actually support the model. Comparisons were made between configurations with and without the mock supports in order to quantify their aerodynamic effects and investigate any changes in flow structure. Force and moment measurements show significant effects on both drag (up to 25 counts / 7% of total drag) and lift (up to 170 counts) due to a vertical strut for all vehicle types. Motor Sport models, whose performance relies greatly on the underside flow, are largely affected (26 counts / 3% on drag and up to 250 counts on lift) by the presence of lateral stings. Passenger vehicle models with larger ride heights were not as sensitive to the use of stings. Further investigation into the flow mechanisms that create these effects were carried out in the form of pressure and velocity measurements in the model and support wakes, surface oil flow visualisations, and surface static pressure readings. Results showed that the strut wake impinged on the rear wings of the motor sport vehicle models and the backlight of the passenger vehicle models as expected, but its influence was more wide ranging than this, extending to the model under floor. To explore changes in flow structure local to the strut-model junction, the junction is simplified as an aerofoil intersecting a flat plate and modelled in Fluent. Comparisons were made between configurations with and without the presence of six different aerofoil profiles for four different boundary layer thicknesses. Results found a noticeable interference on the plate from the union of the aerofoil, but showed that when the magnitude of the interference effect was recalculated using model frontal area the portion of the interference local to the junction affecting the model was small and in some cases insignificant. It was determined that deficiencies in the wake of the supports and their complex interference flow fields were creating a much greater amount of the overall effect by interfering with vehicle features downstream. Due to the complexity of interactions between struts, stings, and the model, the effect of combining them for a common moving ground configuration was highly vehicle dependent, precluding the possibility of developing a reliable correction for interference effects. The results do, however, lead to suggestions for support-model coupling methods that minimise the magnitude of the effect and offer guidelines for the expected magnitudes of effects for the different vehicle types and struts tested.