Validation Case: DrivAer Model Fastback Smooth Underbody

This validation case falls under the domain of fluid mechanics, specifically addressing the aerodynamic characteristics of the Fastback (smooth underbody) DrivAer car model. The objective of this study is to assess and validate the following parameters by employing the Incompressible solver in SimScale:

Drag coefficient ($C_d$)
Pressure coefficient ($C_p$)

The simulation results from SimScale were compared to the experimental data presented in the study “EXPERIMENTAL AND NUMERICAL INVESTIGATION OF THE DRIVAER MODEL”$^1$.

View Project

Geometry

As mentioned before, this validation case uses the Fastback (smooth underbody) car model.

Fastback smooth undetbody drivaer car model that was used in the validation case — Figure 1: Different views of the Fastback smooth underbody DrivAer car model that was used in the validation case

The car model has a reference length of 4.6 $m$ and a reference frontal area of 2.16 $m^2$. A flow volume was created around the car to model a virtual wind tunnel in the SimScale Workbench, as shown below:

enclosure region around the car that will act as a virtual wind tunnel in the simulation — Figure 2: Flow volume around the car model acts as a virtual wind tunnel.

The dimensions of the wind tunnel can be seen in the table below:

Length L $[m]$	Width W $[m]$	Height H $[m]$	Blockage Ratio $[\phi]$ [%]
48	20	12	1

Table 1: Dimensions of the wind tunnel

Did you know?

In wind tunnel testing, the blockage ratio quantifies the degree to which a test model obstructs the wind tunnel’s test section. It is calculated as the ratio of the model’s projected frontal area to the test section’s cross-sectional area, usually expressed as a percentage. A higher blockage ratio leads to greater flow acceleration and pressure changes, affecting the accuracy of the measurements, and may necessitate the use of blockage correction factors.

Analysis Type and Mesh

Tool Type: OpenFOAM®

Analysis Type: Steady-state, Incompressible flow with k-omega SST turbulence model

Mesh and Element Types:

The Standard Mesher algorithm with tetrahedral, hexahedral, and pyramidal cells was used to generate the mesh, with refinements near the walls and in the wake region (see Figure 3).

The details of the mesh for the car model can be seen in the table below:

Mesh	Refinements	Target size $[m]$	Number of Cells
Standard-Automatic	Car surface Near-car region Distant-car region	0.007 0.04 0.08	27 055 957

Table 2: Mesh details for the Fastback car model that was used for validation

In total, three refinements were used to capture the surface profile as well as the wake accurately. The mesh generated is as follows:

generated mesh of the fastback standard mesher — Figure 3: Generated mesh of DrivAer car model. The surface and the two region refinements are clearly visible.

Simulation Setup

Fluid:

Air
- Kinematic viscosity $(\nu)$: 1.567e-5 $m^2/s$
- Density $(\rho)$: 1.196 $kg/m^3$
- Reynolds number $(Re)$: 4.87e6

Boundary Conditions

The boundary conditions for the simulation are shown in Figure 4 below:

Boundary Condition	Face	Value
Velocity inlet – Fixed value	Inlet	16 $[m/s]$
Pressure outlet – (Fixed value, gauge)	Outlet	0 $[Pa]$
Slip wall	Side and top faces	–
Moving wall – Wall function	Bottom face (Ground)	16 $[m/s]$
Rotating wall – Wall function	Wheels	51 $[rad/s]$ (angular velocity)

Table 3: Detailed boundary conditions applied on the fluid domain

Reference Solution

This case is a validation against the reference result obtained through an experiment where the Fastback DrivAer car model is placed in a wind tunnel as described in Figure 4 of the study $^1$. The car model has a scale of 1:2.5, and the dimensions of the wind tunnel are:

L $[m]$	W $[m]$	H $[m]$	Blockage ratio $[\phi]$ [%]
4.8	2.4	1.8	8

Table 4: Dimensions of the Wind Tunnel A used in the experiment (see Figure 4 in the study $^1$)

Result Comparison

As mentioned before, this validation case compares the pressure and drag coefficient obtained from SimScale against the experimental results obtained from the study referenced$^1$.

1. Drag Coefficient ($C_d$)

The drag coefficient was directly obtained from the SimScale Workbench using the result control plots.

drag coefficient plot fastback drivaer car — Figure 4: Converged drag coefficient plot obtained in SimScale Workbench

The calculated drag coefficient from the simulation was then compared with the experimental results, which can be seen in the table below:

Car Model	Experimental	Simulation	Error [%]
Fastback – smooth Underbody	0.243	0.254	4.5

Table 5: Drag coefficient comparison between the experiment and the simulation

2. Pressure Coefficient ($C_p$)

To calculate the pressure coefficient, the pressure values at the top and bottom of the car surface along a centerplane normal to the y-axis need to be extracted. This was done using ParaView. With the obtained pressure value, the pressure coefficient was calculated with the formula below:

$$C_p = \frac{p-p_{\infty}}{\frac{1}{2}\rho\ U^2}\tag{1}$$

where:

$p$: static pressure at the point of calculation $(Pa)$
$p_{\infty}$: static pressure in the freestream $(Pa)$
$\rho$: freestream fluid density $(kg/m^3)$
$U$: freestream fluid velocity $(m/s)$

The comparison of pressure coefficients for the car model along the centerplane normal to the y-axis at the top and bottom can be seen in the figures below:

pressure coefficient at the top surface of fastback drivaer car model obtained from the simulation and experiment to use for comparison — Figure 5: Pressure coefficient comparison at the top surface of the car along the centerline, for the fastback smooth underbody

pressure coefficient at the bottom surface of fastback drivaer car model obtained from the simulation and experiment to use for comparison — Figure 6: Pressure coefficient comparison at the bottom surface of the car along the centerline, for the fastback smooth underbody

Furthermore, a detailed comparison of the pressure coefficient on the front and rear windshields and the side window of the car model can be observed in the figures below:

Figure 7: Visual comparison of pressure coefficient on the front windshield of the Fastback model, (left) Experiment and (right) SimScale

Figure 8: Visual comparison of pressure coefficient on the side window of the Fastback model, (left) Experiment and (right) SimScale

Figure 9: Visual comparison of pressure coefficient on the rear windshield of the Fastback model, (left) Experiment and (right) SimScale

The pressure coefficient distribution around the complete fastback model as observed from simulations:

visualization of pressure coefficient for the fastback car model — Figure 10: Pressure coefficient over the complete fastback car model

References

Heft, Angelina I., Thomas Indinger, and Nikolaus A. Adams. 2012. “Experimental and Numerical Investigation of the DrivAer Model.” In Proceedings of the ASME 2012 Fluids Engineering Summer Meeting (FEDSM2012-72272), July 8–12, Rio Grande, Puerto Rico. New York: American Society of Mechanical Engineers.

Last updated: September 28th, 2025

Did this article solve your issue?

How can we do better?

We appreciate and value your feedback.

Length L \([m]\)	Width W \([m]\)	Height H \([m]\)	Blockage Ratio \([\phi]\) [%]
48	20	12	1

L \([m]\)	W \([m]\)	H \([m]\)	Blockage ratio \([\phi]\) [%]
4.8	2.4	1.8	8