13-09-2014, 02:12 PM
CFD Simulation of Pickup Van
CFD Simulation.pdf (Size: 1.53 MB / Downloads: 36)
ABSTRACT
Broadly, the objective of this study is to
carry out 3-Dimensional, incompressible,
steady-state CFD simulation of a simplified
pickup van with smooth underbody and without
side mirrors, by employing pressure based
commercial software CFDExpert-LiteTM to
carry out Reynolds Averaged Navier-Stokes
(RANS) based computations to investigate the
aerodynamics of pickup van. k-ε turbulence
model with standard wall functions and
structured CFD domain is used in the
simulation to study the flow parameters in and
around the wake region and a detailed study of
the quantitative data set for validation of
numerical simulations has been conducted.
Simulations were carried out at moderate
Reynolds numbers (~3×105
) and the measured
quantities include: the pressure distributions on
the symmetry plane and the velocity profiles
near the wake.
INTRODUCTION
Pickup vans are one of the more popular vehicles
in use today yet it has received very little attention in car
aerodynamics literature. The aerodynamics of pickup vans
is more complex than any other open bed trucks, because
the short length of the bed can result in interaction of the
bed walls and tailgate with the separated shear layer
formed at the edge of the cab. The present study is to gain
a better understanding of the flow structure near the wake
region, since the theories on the aerodynamics are yet to
mature and wind tunnel experiments cost long periods and
great expenses, the numerical simulation based on
computational fluid dynamics (CFD) is a good approach
to adopt.
METHODOLOGY
Vehicle Geometry
The model represents a 1:12 scale of the full-scale
pickup truck. Schematics of the pickup truck model with
the pertinent dimensions are depicted in Figure 2.1. The
length of the model is 0.432 m, the width is 0.152 m, and
the height is 0.1488 m [1] [5]. The model was designed
with a smooth underbody, enclosed wheel-wells and
without openings for cooling airflow. An identical surface
model was generated for the CFD simulations. Once the
surfaces for the model are created, they are used to
generate a three-dimensional grid for the CFD
calculations.
Solver:
In the present study all the computations for the
computational domain have been carried out using a three
dimensional RANS model with an industry standard finite
volume based CFD code, CFDExpert-Lite. The set of
equations solved by CFDExpert-Lite are the unsteady
Turbulence Modeling
The standard
k
model is used for the simulations.
Since the boundary condition for
k
is not well
defined near the wall, one uses the law of the wall as the
relation between velocity and surface shear stress.
Evaluation of shear stress depends on whether the near
wall cell lies in the viscous sub layer or in the fully
turbulent region as decided by non-dimensional distance
CONCLUSIONS
An investigation of the flow in the near wake of a pickup
truck model has been conducted. The main conclusions of
the investigation are:
Mean pressure data show the expected behavior
at the front of cab, and a cab base pressure
coefficient in the range Cp ~ -0.25 to -0.35.
The mean pressure distribution on the tailgate
show a lower pressure coefficient on the inside
surface compared to the outside surface
suggesting that the tailgate reduces aerodynamic
drag.
The pressure fluctuations at the cab base are very
low and increases significantly towards the back
of the bed and the tailgate top edge.
Mean velocity field measurements in the
symmetry plane show a recirculating flow region
over the bed bounded by the cab shear layer. The cab shear layer does not interact directly with the
tailgate..
The underbody flow results in the formation of a
strong shear layer in the near wake.
One of the more striking features of the pickup
truck flow is the downwash on the symmetry
plane behind the tailgate, and the formation of
two smaller recirculating flow regions on both
sides of the symmetry plane. These features are
consistent with the formation of streamwise
vortices in the wake.
An approach of this nature would greatly reduce
design time and make CFD a more feasible
option.