Graham Doig

Influence of wing span on the aerodynamics of wings in ground effect

A computational fluid dynamics study of the influence of wing span has been conducted for an inverted wing with endplates in ground effect. Aerodynamic coefficients were determined for different spans at different ground clearances, highlighting a trend for shorter spans to delay the onset of both separation and resulting loss of negative lift. The vortices at the wing endplates were not observed to change significantly in terms of strength and size; thus, at shorter spans, their influence over a larger percentage of the wing helped the flow stay attached and reduced the severity of the adverse pressure gradient which invokes separation at greater spans. Consequently, it was shown that, compared to a large-span wing, a wing with a shorter span may have a lower lift coefficient but can operate closer to the ground before performance is adversely affected.

Flow compressibility effects around an open-wheel racing car

A numerical investigation has been conducted into the influence offlow compressibility effects around an open-wheeled racing car. A geometry was created to comply with 2012 F 1 regulations. Incompressible and compressible CFD simulations were compared- firstly with models which maintained Reynolds number as Mach number increased, and secondly allowing Mach number and Reynolds number to increase together as they would on track. Results demonstrated significant changes to predicted aerodynamic performance even below Mach 0·15. While the full car coeffixad cients differed by a few percent, individual components (particularly the rear wheels and the floor/ diffuser area) showed discrepancies ofover 10% at higher Mach numbers when compressible and incompressible predictions were compared. Components in close ground proximity were most affected due to the ground effect. The additional computational expense required for the more physically-realistic compressible simulations would therefore be an additional consideration when seeking to obtain maximum accuracy even at low freestream Mach numbers.

Synergistic integration of computational fluid dynamics and experimental fluid dynamics for ground effect aerodynamics studies

This article highlights the ‘synergistic’ use of experimental fluid dynamics (EFD) and computational fluid dynamics (CFD), where the two sets of simulations are performed concurrently and by the same researcher. In particular, examples from the area of ground effect aerodynamics are discussed, where the major facility used was also designed through a combination of CFD and EFD. Three examples are than outlined, to demonstrate the insight that can be obtained from the integration of CFD and EFD studies. The case studies are the study of dimple flow (to enhance aerodynamic performance), the analysis of a Formula-style front wing and wheel, and the study of compressible flow ground effect aerodynamics. In many instances, CFD has been used to not only provide complementary information to an experimental study, but to design the experiments. Laser-based, non-intrusive experimental techniques were used to provide an excellent complement to CFD. The large datasets found from both experimental and numerical simulations have required a new methodology to correlate the information; a new post-processing method has been developed, making use of the kriging and co-kriging estimators, to develop correlations between the often disparate data types.

Aerodynamic design and development of the Sunswift IV solar racing car

The aerodynamic design and development of the University of New South Wales’ ultra-low-drag solar-electric Sunswift IV car is described, detailing the student-led design process from initial concept sketches to the completed vehicle. The body shape was established and refined over a period of six months in 2008–2009, almost entirely using computational fluid dynamics. The guiding philosophy was that predictable handling and drag minimisation in challenging, changing wind conditions of the type commonly seen during the World Solar Challenge across Australia was preferable to high performance only on ‘perfect’ days. The car won its class in the 2009 and 2011 World Solar Challenges, and holds the Guinness World Record for fastest solar-powered vehicle.

Interaction of shock tube exhaust flow with a non-pre-mixed flame

Graphical abstract

The aerodynamics of a supersonic projectile in ground effect

A computational aerodynamic study of a bullet travelling in ground effect at Mach 2.4 has been conducted, based on physical experiments conducted by UNSW@ADFA in Canberra, Australia. The investigation aimed to establish the way in which the aerodynamic characteristics of the bullet change due to close proximity to the ground. CFD was validated against the experimentally observed flow field and pressure measurements and was subsequently used to predict the aerodynamic forces on the projectile.

The Aerodynamics of a Cornering Inverted Wing in Ground Effect

For racing car configurations an inverted wing produces negative lift that allows increased levels of acceleration to be maintained through corners. Routine aerodynamic analysis, however, will typically be in the straight-line condition. A numerical analysis of the inverted T026 wing geometry through the curved path of a constant radius corner was conducted. The asymmetrical properties of the oncoming flow resulted in the introduction of a rolling and yawing moment along the span, as well as side-force. Yaw angle, flow curvature and a velocity gradient resulted in changes to the pressure distribution over the wing surface. Primary vortex behaviour was observed to differ significantly in both direction and structure.

Application of Kriging to Motorsport Aerodynamic Analysis

Aerodynamic analysis in motorsport is conducted using three methods, computational, scaled experimental and full-scale operational. However, the varying fidelity, different sampling resolutions and unavoidable errors of each technique make valid comparisons between datasets from each method difficult and time consuming. Kriging is a geostatistical method to estimate values within a data field by examining and applying the trends of the dataset. This research examines how such techniques can be used to aid comparison between aerodynamic measurements of a race car. It examines how kriging can be used to transform discrete measurements, of varying fidelity and sampling resolution, into semi-continuous measurements, thus allowing computational results to be compared across a wider range of conditions than initially tested. This work explores how kriging can allow the trends from highly sampled data, such as track running, to be applied to less sampled data, such as CFD to improve computational and overall aerodynamic analysis.

On the Interaction of a Racing Car Front Wing and Exposed Wheel

A numerical investigation of generic open-wheel racing car wing and wheel geometry has been conducted, using original sub-scale experimental data for validation. It was determined that there are three main interactions that may occur, identifiable by the path that the main and secondary wing vortices take around the wheel. Interaction ‘A’ occurs when the main and secondary wing vortices both travel outboard of the wheel; interaction ‘B’ is obtained when only the main wing vortex passes inboard of the wheel; while interaction ‘C’ sees both wing vortices travel inboard of the wheel. The different interactions are achieved when geometric changes to the wing affect the pressure distribution about the endplate, either by altering the magnitude of suction generated by the wing or by changing the locations of peak suction and vortices relative to the wheel’s stagnation regions. As a result, the influence that the wing and wheel have on each other – in comparison to the same bodies in isolation – varies, resulting in significant consequences for downforce and drag.

Implications of compressibility effects for Reynolds-scaled testing of an inverted wing in ground effect

The influence of compressibility around an isolated inverted wing at a fixed Reynolds number was examined as relevant to the issue of wind tunnel scaling effects. Three-dimensional simulations were conducted for low ground clearances, at: full scale and a Mach number of 0.088, at 50% scale at Mach 0.176, and at 25% scale at Mach 0.352. As the scale was reduced, the increasing peak local Mach number between the wing and the ground resulted in a higher propensity of the flow to separate towards the trailing edge, and for incompressible or full-scale CFD to underestimate the lift and drag coefficients by an ever-increasing margin. The lower vortex path was less affected. The results suggest that compressible CFD of a scale experiment ought to be conducted at the same Reynolds number and Mach number as the tunnel test for the best possible correlation at free-stream Mach numbers beyond 0.15.