Test Data AC1-05
Ahmed body
Application Challenge 1-05 © copyright ERCOFTAC 2004
Overview of Tests
The first experiment (Exp1) deals with conventional average values such as oil flow patterns, static pressure measured on large times. Thus no information about unsteadiness will be included in the data. In Exp1, conventional wake surveys are performed with 10 hole probes. Drag is measured by strain gauge balance. The contribution of the drag is estimated for each part of the model. (front, slant rear, vertical rear base). The Reynolds number based on the model total length is 4.29 x 10^{6}.
The second experiment (Exp2) uses a two-components LDV system. Averages are performed on a high number of samples (40000) for long time durations, typically 5 minutes.
NAME | GNDPs | PDPs | MPs | |||
---|---|---|---|---|---|---|
Re | External velocity | External turbulence level | Slant angle | Detailed Data | DOAPs | |
EXP 1 Ahmed original (1984) | 4.29x10^{6} | 60ms^{-1} | 0.5% | 5°, 12.5°, 25°, 30° | P_{w}, U_{i} | C_{x}, Flow structure |
Re | Detailed Data | DOAPs | ||||
EXP2 Lienhart et al. (2000) | 2.78x10^{6} | 40ms^{-1} | 0.25% | 25°, 35° | First, second and third moments | P_{w}, Flow structure |
Test Case EXP-1
Description of Experiment
See table
Boundary Data
The test section is a ¾ open test section. Only the floor is a solid boundary. The homogeneity and far field conditions are not given.
The incoming turbulence intensity is less than 0.5% for 60 ms^{-1}. No details on the incoming turbulence are available.
The size of the nozzle at the entrance of the test section is 3x3 m^{2}.
The model is supposed to be smooth. No info is available on the turbulent/laminar nature of the boundary layers on the model. The influence of possible transition can be tested by CFD;
No information is available on the precision of the alignment of the model in the flow, although the symmetry on the visualizations give some confidence on this point. This sensitivity can also be checked by CFD.
No details on the typical time scales are provided. The influence of the unsteadiness of the wake on the averaging process can be estimated through URANS or LES.
Measurement Errors
Flow angle precision ± 0.4°.
Free stream dynamic pressure 1%.
Forces and moments are measured with balances with uncertainty of
± 0.2 N and ± 0.1Nm.
Measured Data
The test data include measurements of:
- Wall pressure
- Visualization of flow patterns on rear (slant) surface
- Wake survey (velocity vector plots, average values) :
- Mean velocity distribution in wake central plane
- Cross flow velocity for several downstream locations
- Drag coefficient: contributions of the pressure and friction drags to the total drag are estimated, as well as the repartition of the pressure drag among the front, slant part and vertical base.
All these data are provided for slant angles j = 5°, 12.5°, 25° and 30°.
An additional test is performed by fixing a splitter plate vertically in the wake of the body, in the plane of symmetry.
References
Some salient features of the time-averaged ground vehicle wake, S.R. Ahmed, G. Ramm and G. Faltin, SAE paper series Technical paper 840300, Detroit, 1984
Test Case EXP-2
Description of Experiment
See table
Boundary Data
The test section is a ¾ open test section. Only the floor is a solid boundary. The homogeneity and far field conditions are not given. However the blockage is assumed to be less than 4%.
The incoming turbulence intensity is less than 0.25% for 40 ms^{-1} measured by hot wire anemometry 400 mm upstream of the model. The viscosity ratio is about 10.
The models are supposed to be smooth. Transition to turbulence of the boundary layer on the front part is triggered.
No information is available on the accuracy of the alignment of the model in the flow, although the symmetry on the visualizations give some confidence on this point. This sensitivity can also be checked by CFD.
No detail on the typical time scales are provided. The influence of the unsteadiness of the wake on the averaging process can be estimated through URANS or LES.
Measurement Errors
Error on mean velocities is less than 0.005% of local mean in the outer flow. In the wake region the accuracy is assumed to be 1% for mean values and 1.5% for RMS.
Measured Data
LDA measurements of mean velocities: U, V, W, Reynolds stresses and third order moments in some planes for 2 slant angles:
25° slant angle:
planes: Ahmed_25_y=0_global.dat (whole flow); Ahmed_25_y=0_focus.dat (focus on the slant part); y=100; y=180; y=195; y=-195.dat x=-178; x=-138;
x=-88; x=-38; x=0; x=80; x=200; x=500
35° slant angle:
planes: Ahmed_35_y=0_global.dat (whole flow); Ahmed_35_y=0_focus.dat (focus on the slant part); y=100; y=180
x=-88; x=0; x=80; x=200; x=500
Hot wire measurements in the boundary layer in the symmetry plane at different x-location: mean velocities, Reynolds stresses and third moments (only u-w components):
25° slant angle:x= -243, -223, -203, -183, -163, -143, -123, -103, -83, -63, -43, -23, -3
35° slant angle:x= -243, -223, -203, -183, -163, -143, -123, -103, -83, -63, -43, -23, -3
Pressure coefficients on the rear of the body:
References
Flow and Turbulence Structures in the Wake of a Simplified Car Model (Ahmed model),
H. Lienhart, C. Stoots and S. Becker, DGLR Fach Symp. Der AG STAB, Stuttgart University, 15-17 nov. 2000
H. Lienhart and S. Becker, Flow and turbulence structures in the wake of a simplified car model, SAE Paper 2003-01-0656, 2003.
© copyright ERCOFTAC 2004
Contributors: Remi Manceau; Jean-Paul Bonnet - Université de Poitiers
Site Design and Implementation: Atkins and UniS