All computations used for this report have been made with Fluent 5.4, with a car velocity of 108 km/h, a k-epsilon turbulence model, with a second order scheme for pressure and momentum. The Reynolds number associated is 3,000,000.
We also sucessfully made a test run using Star-CD.
The following tables report all the computed aerodynamic values, for the three different meshes:
The geometry has been defined with two separate zone:
car and glasses. Car roughly regroups all the steel part of the car body,
while glasses contains all the windows. The board has its own zone (named....
board). Net is the sum of all the forces.
Car without board
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The car has got a coefficient Cx (or Cd) about 0.30, which
is very low if compared to real vehicles (about 0.35 to 0.40; VW Polo:
0.37). This value is acceptable if we consider that this car has been designed
without wheels or accessories such as rear view mirrors, calender, handles,
windscreen wipers...
For exemple, rear view mirrors can account for as much
as 5% of the drag.
The main contribution in the total force Fx is due to
pressure effects, much greater than viscous effects.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Car with board in first configuration
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The computed drag coefficient is now about 0.346. The
efforts on the board are not significant compared to car forces, but the
presence of this new object on the roof deeply modifies the flow around
the car. That's why forces on the car have increased.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Car with board in second configuration
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The drag coefficient has now increased until 0.385. The major effect of this configuration is that it creates a great overpressure on the windscreen (twice more important than with the first configuration). This overpressure can be seen on the contours of pressure displayed.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The underpressure zone under the board is then larger than for the previous configuration, and the force on the board is then greater. See contours of pressure.
The main results are reported in the following table:
|
|
|
|
|
|
|
|
|
|
|
|
Conclusion:
The first configuration is consequently much more economic
than the second one. In conclusion, it seems better to put the board nose
forward.
Nevertheless, it makes the drag coefficient increase,
compared to the same car without anything on its roof.