# Virtual Wind Tunnel and Free Air CFD Comparison Part 2/2

Welcome to the concluding episode of my two-part project to compare results between the virtual wind tunnel and free air using a series of Computational Fluid Dynamics (CFD) simulations. Recall that the previous post outlined the free air tests and expectations for the comparison. Here I present the results and conclusions.

Pressure contours on the racecar

### Tests

I ran three CFD simulations in free air to compare with the wind tunnel results:

- 40% scale model in free air matching the wind tunnel conditions, Re = 3x10
^{6}, U = 29 m/s - Full size model in free air with matching Reynolds number to wind tunnel, Re = 3x10
^{6}, U = 12 m/s - Full size model in free air with matching velocity to wind tunnel, Re = 7.5x10
^{6}, U = 29 m/s

### Results

Pressure contours on the racecar

The reference area for the drag and lift coefficients was half the area of the plane that fits the maximum height and width of the racecar.

#### Drag of Racecar and Wheels

Relative to wind tunnel drag coefficient

All the drag coefficient values are within 5% to 6.5% of the wind tunnel (WT) value. As expected there is very little difference in drag coefficients between the free air results for the 40% model and the full size model with matching Reynolds number.

The fact that the drag coefficient for case 3, with a higher Reynolds number, is close to that of case 2 means that the air flow around the racecar is not particularly sensitive to Reynolds number, at least for drag. This bodes well for using the wind tunnel to guide the full size aerodynamic racecar design, however, before we can make that judgement we need to examine the lift coefficients.

#### Drag of Racecar Only

Relative to wind tunnel drag coefficient

Excluding the drag from the wheels to focus on the racecar shows that the wheels account for approximately 50% of the overall drag for all cases.

Also we can see that the drag coefficients for the free air cases with matching Reynolds numbers are reasonably close to the equivalent value in the wind tunnel when the wheel drag is excluded. Case 3, with a higher Reynolds number, is the furthest away from the wind tunnel drag coefficient with a difference of 12.2%.

#### Lift of Racecar and Wheels

Relative to wind tunnel lift coefficient

There are relatively large discrepancies between the wind tunnel lift coefficient (negative values indicate downforce) and the equivalent values for the free air cases, i.e., 17.8% to 23.2%. This indicates that the wind tunnel has a larger effect on the lift than it does on the drag.

As expected there is very little difference in lift coefficient between the free air results for the 40% model and the full size model with matching Reynolds number. Also the relatively small variations in lift coefficients between the full size free air cases confirms that this configuration is not particularly sensitive to Reynolds number.

#### Lift of Racecar Only

Relative to wind tunnel lift coefficient

Excluding the lift from the wheels to focus on the racecar shows that the wheels reduce the effective downforce by approximately 50%.

Also we can see that the lift coefficients for the free air cases are reasonably close to the equivalent value in the wind tunnel when the wheel lift is excluded. Interestingly case 3, with a higher Reynolds number, is the closest to the wind tunnel lift coefficient with a difference of only 4.8%.

### Conclusions

Clearly there are differences between the free air cases and the wind tunnel results as shown by the drag and lift comparisons. To get better agreement between the wind tunnel and free air experiments there are a variety of options to consider.

In actual wind tunnels the boundary layer that forms upstream of the rolling road is often removed by suction, which then better mimics the actual movement of the racecar close to the ground. Also many wind tunnels with rolling roads isolate the aerodynamic forces acting on the wheels by not directly mounting them on the racecar. Boundary layer suction and isolated wheel mounts are employed by the physical wind tunnel that formed the basis for this virtual wind tunnel study. Also wind tunnel wall corrections are applied to wind tunnel results to compensate for the proximity of the walls relative to the test model.

The primary consideration for racecar success is downforce. Isolating the downforce (lift) on the racecar from the wheels, as presented in the "Lift of Racecar Only" Results section, shows that the wind tunnel can produce reasonable data to guide the aerodynamic design of racecars, even without matching the Reynolds number. This is not lost on leading motorsport teams that make extensive use of wind tunnels to assess and optimize their racecars.

## Trending Now

## Select Language

## Recent blog posts

- External Aerodynamics with CFD
- A Case for Renaming the Navier-Stokes Equations
- Ludwig Prandtl: Real Fluids Explained
- Osborne Reynolds: A Giant in Fluid Dynamics
- Sliver Treatment Strategies for CFD
- How to Fix Small Acute Angles for CFD
- Small Feature Removal for CFD
- Fluid Visualization in Nature
- Design is Compromise
- CFD Is Not Enough