Application Challenge 1-05 © copyright ERCOFTAC 2004
A basic ground vehicle type of bluff body is investigated. The body consists of three parts : a fore-body, a mid section and a rear end.
Two experiments are available:
The first one (Exp1) was performed at DLR-Göttingen in a wind tunnel at Reynolds number 4.29 million (60 m/s), based on the model length. The model is mounted on a ground plate, in order to reproduce the ground effect. The angle of the rear end slope is adjustable, between 0 and 40° with a 5° step. More details are available for angles of 5°, 12.5° and 30°. Pressure is measured by about 210 pressure probes on the fore-body, 83 in the mid section and 450 on rear ends. Friction lines visualizations are also available. Moreover, detailed wake surveys are performed with 10 hole probes and drag measurements are provided.
The second, more recent experiment (Exp2) was provided by Erlangen LSTM within the MOVA “Models for Vehicle Aerodynamics” European project (TU Delft, Univ. of Manchester (UMIST), LSTM, Electricité de France, AVL List, PSA Peugeot-Citroën). The same model is used as in the previous study, but the Reynolds number is reduced to 2.78 million (40m/s), and the study is focused on slant angles close to the drag crisis, 25° and 35°. Two-component hot wire measurements were performed in the boundary layer above the slant part and LDA measurements in 13 different planes. Mean values and turbulence statistics (second and third moments) are provided. Pressure measurements were performed on the rear part of the model (435 pressure probes). Oil/soot friction lines visualizations are also provided.
The Ahmed body was one of the test cases of the 9th and 10th ERCOFTAC-IAHR Workshop on Refined Turbulence Modeling held in Darmstad, Germany (2001) and Poitiers, France (2002) respectively. These workshops were organized under the auspices of the Special Interest Group 15 on Turbulence Modeling of ERCOFTAC. The proceedings of the 10th ERCOFTAC-IAHR Workshop can be found at:
and this test case at:
CFD results were obtained by 15 different teams, ranging from simple RANS models (standard k-epsilon model with wall functions) to more elaborate RANS models and even LES.
Relevance to Industrial Sector
The basic shape of the so-called « Ahmed body » contains important features of real road vehicles : a main body, followed by a fully separated region. Prediction of separated flows is one of the more difficult task of CFD. The parametric variation of the angle of the rear part allows the study of various configuration relevant to real car characteristics, from massively separated, “simple” wakes, to very complex, 3D wake structures. The reproduction of this complex, 3D wake is very challenging for CFD, as well as the transition from one behaviour to another. The data base contains also drag results that are essential to predict for practical purposes, and are closely related to the structure of the wake.
The second set of experiments provides very detailed results, including turbulent quantities that are useful for a detailed analysis of turbulence models.
Design or Assessment Parameters
The first DOAP is the drag coefficient, and in particular its variations with the slant angle.
A second DOAP is the topology of the flow, which is crucial for the correct reproduction of the drag coefficient. Comparisons between computations and experiments in the provided planes and in the boundary layer will be useful, as well as friction lines visualizations on the slant part and the vertical base.
Flow Domain Geometry
The model is described on Figure 1. The geometry of the fore-body is available in Ahmed_Front_Geo.dat.
X, Y and Z are the streamwise, spanwise and ground-normal directions, respectively. The origin of the axes is the point at the intersection between the vertical base (X=0), the symmetry plane (Y=0) and the ground plate (Z=0).
Overall length: 1.044 m. Width 0.389 m, Height 0.288 m.
The forebody is 0.182 m long, the center of the curvature being placed 100 mm from the front and upper/lower/lateral surfaces. The central (constant section area) is 0.640 m long.
All rear ends have the same slant part length Ls= 222mm. The edges are sharp.
The model is placed in a 3/4-open test section (only the floor is present). No indication on the homogeneity of the flow is given.
Special attention should be given on the presence of a ground plate between the tunnel floor and the body. This plate is used for preventing wind tunnel boundary layer parasitic effects on the model. The model lies 50 mm above it and is placed on stilts of 30 mm diameter.
Flow Physics and Fluid Dynamics Data
The flow has no special characteristics. Air at ambiant conditions is used. The model is supposed to be smooth.
© copyright ERCOFTAC 2004
Contributors: Remi Manceau; Jean-Paul Bonnet - Université de Poitiers
Site Design and Implementation: Atkins and UniS