A correlation study of wind tunnels for reduced-scale automotive aerodynamic development.
Wind tunnel testing of reduced-scale models is a valuable tool for aerodynamic development during the early stages of a new vehicle program, when basic design themes are being evaluated. Both full-and reduced-scale testing have been conducted for many years at the General Motors Aerodynamics Laboratory (GMAL), but with increased emphasis on aerodynamic drag reduction, it was necessary to identify additional facilities to provide increased test capacity. With vehicle development distributed among engineering teams around the world, it was also necessary to identify facilities local to those teams, to support their work. This paper describes a cooperative effort to determine the correlation among five wind tunnels: GMAL, the Glenn L. Martin Wind Tunnel (GLMWT) at the University of Maryland, the Institute of Aeronautics and Space (IAE/ALA) TA-2 Wind Tunnel in Brazil, the Monash University Wind Tunnel in Australia, and the Korea Aerospace Research Institute Low Speed Wind Tunnel (KARI LSWT). Correlation tests were conducted using a vehicle model with interchangeable body modules and additional parts, to determine the relationships of measured aerodynamic force and moment coefficients among the tunnels. Despite the significant differences among facilities, the correlation results were satisfactory for reduced-scale wind tunnel development of future General Motors Co. (GM) vehicles. Correlation equations were defined, allowing conversion of data from one tunnel to equivalent values at another. This paper will present the physical descriptions and airflow characteristics of the wind tunnels, the test equipment and procedures, and the correlation results.
CITATION: Meinert, F., Johannessen, K., Saito, F., Song, B. et al., "A Correlation Study of Wind Tunnels for Reduced-Scale Automotive Aerodynamic Development," SAE Int. J. Passeng. Cars - Mech. Syst. 9(2):2016.
Wind tunnel testing of reduced-scale models has long been recognized as an efficient tool for aerodynamic shape development in the aircraft and automobile industries, among others . At GM, reduced-scale models are used during the early stages of a new vehicle development program, when significant changes in design themes and features are evaluated. It is simply easier and faster to make large changes on a small model.
Increasingly stringent global requirements for fuel economy and C[O.sub.2] emissions require ever lower aerodynamic drag, which in turn drives additional development work, even with the continuing improvement of analysis tools. Consequently, it became necessary for GM to identify and qualify additional wind tunnel facilities as the demand for test time exceeded the available capacity at GMAL. In addition, GM's future product development programs are spread among regional technical centers around the world, and the ability to perform development in nearby wind tunnels would greatly improve the efficiency of the process.
For these reasons, wind tunnels suitable for testing 1/3-scale models were identified in the United States, Brazil, Korea, and Australia. A series of correlation tests was conducted, using a model with interchangeable rear body modules that provided a variety of basic shapes together with additional parts, which produced a large number of individual configurations and their associated increments. Correlation equations based on this data set provide common, comparable measures of aerodynamic performance regardless of the source. The following sections will describe the wind tunnels, the project approach and observations, and the comparative results.
General Motors Aerodynamics Laboratory (GMAL)
GMAL was built during 1977-1980 at the GM Technical Center in Warren, Michigan, USA . It was intended primarily for aerodynamic shape development of full-scale cars and trucks. A significant fraction of the operation is also dedicated to aero-acoustic testing, with that capability enhanced by extensive acoustic modifications in 2001 and 2006. Various motorsports and nonautomotive tests have added variety to the everyday operations, but those have composed a very small portion of the total usage. Major upgrades of the data acquisition and control system were performed approximately every ten years, with the most recent in 2010.
GMAL is a large wind tunnel with a closed-jet test section measuring 5.44 m high x 10.36 m wide. For full-scale cars and light trucks, this produces a blockage ratio of 6% or less. It should be noted that the blockage with a 1/3-scale model in the other wind tunnels described herein is of similar magnitude to the full-scale blockage in GMAL. The test section has a stationary floor with a boundary layer suction scoop at its entrance. A secondary distributed suction panel just ahead of the full-scale balance may be used to minimize the boundary layer thickness for especially low vehicles. A smaller balance at the front of the test section enables development of 1/3-scale models. This provides a fairly unique testing environment with extremely low blockage of the test section (<0.7%). The reduced-scale balance is located just behind the suction scoop to take advantage of the thin boundary layer. It is an external (under-floor), six-component, strain gage balance. Mounting studs that support either the wheels or axles of the model are adjustable for a range of track and wheelbase dimensions by rotating eccentric circles in the turntable surface.
Glenn L. Martin Wind Tunnel (GLMWT) - University of Maryland College Park
The GLMWT has been actively involved in aerodynamic research and development since 1949. It was constructed as part of a gift to the University in the late 1940s. A very broad range of subjects have been tested, with over 1800 tests to date . Frequent upgrades have maintained its state of the art capabilities. The tunnel has an external, six-component balance with structure under and alongside the test section, permitting the measurement of forces and moments on models mounted on the floor or in the center of the test section.
The test section is of closed-jet design, and for automotive testing a secondary floor assembly is installed with provision for boundary layer suction and a model support structure enclosed in a turntable. A unique feature of the GLMWT compared to the other wind tunnels described in this paper is the model mounting hardware. Four circular pads of approximately 150 mm diameter support the model wheels, and are exposed to the airflow. Pressure taps in the pads allow for subtraction of the aerodynamic loads on the pads.
Korea Aerospace Research Institute Low Speed Wind Tunnel (KARI LSWT)
The KARI Low Speed Wind Tunnel was developed in 1998 for the primary purpose of supporting aeronautical and aerospace development programs. However, the facility was also designed to support other test objectives, including ground vehicles . The primary, closed-jet test section with 3 m x 4 m rectangular cross-section was installed in 1998. The maximum wind speed in the closed test section is 120 m/s and the turbulence intensity level is less than 0.13%. The alternate, open-jet test section with a 3.75 m x 5 m nozzle was installed in 2008. The maximum wind speed in the open jet test section is 70 m/s and the turbulence intensity level is less than 0.4%.
More than 200 models such as a fixed wing airplane, helicopter, launching vehicle, high-speed train, wind turbine blade, ship, and motor vehicle were tested in this facility since 1998. Most of the tests focused on the aerodynamic forces and moments and surface pressures. Some of the tests were conducted to measure the aeroacoustic noise and flow field around the model. The six-component balance, electronic pressure scanner, laser beam, and acoustic measuring devices were installed to support those kinds of tests.
An elevated ground board system was installed in 2010 to reduce the boundary layer thickness for simulation of the ground effect of airplanes. The total length including the leading edge and tail flap was 6.5 m and the height from the tunnel floor to its center was 0.85 m. It was modified in 2013 for automotive testing. An automotive support frame was designed and built with attachment directly to the external balance. Four adjustable wheel supports permit various sizes of models to be mounted. The front wheel position can be adjusted from 2.5 m to 2.7 m behind the leading edge of the ground board.
Monash University 1.4 MW Wind Tunnel
The Monash 1.4 MW Wind Tunnel is the largest wind tunnel in Australia. It is a multi-purpose wind tunnel that is capable of full-scale vehicle aerodynamic and aero-acoustic testing. It is also used for a variety of wind engineering studies and for applications such as aircraft, truck, train, cycle, and other aerodynamic tests. A schematic of the tunnel's layout is shown in Figure 8. The tunnel is a closed return design, where the air path follows a vertical circuit. In standard configuration it has two test sections: a 2.6 m x 4 m 3/4-openjet automotive test section (downstairs) and a 4 m x 12 m closed-jet wind engineering section (upstairs). By raising the top of the nozzle, the automotive test section can be increased to 4 m x 4 m. The air is driven by two 5 m diameter fans, while a number of acoustic splitters and dampeners are employed to reduce the noise level in the open-jet test section for aero-acoustic testing.
A project was undertaken to develop a new closed test section that could be inserted into the automotive test section to provide the capability to conduct 1/3-scale vehicle aerodynamic tests. The design of this new scale model test section needed to incorporate a new scale model force balance and turntable, and its geometry had to be carefully designed to ensure adequate flow uniformity, minimal static pressure gradient, and minimal boundary layer height at the model.
The new scale test section was designed to be installed into the wind tunnel's open-jet automotive test section as a closed-jet configuration. The walls of the insert attach to the nozzle exit of the existing open-jet automotive section and extend 10.8 m downstream to the collector. The insert needed to be removable so that the wind tunnel could be converted back to its full-scale open-jet configuration. Therefore the walls were designed as modular sections that are bolted together to form the closed-jet walls.
An elevated ground plane design was used to minimize boundary layer effects at the model. A CFD-optimized nose profile was developed with the final dimensions for the ground plane being 8.8 m long with the leading edge 2.8 m upstream of the turntable center. A half ellipse profile was selected for the nose geometry and the ground plane was tapered over the last 1.2 m at the trailing edge. The ground plane is supported by ten 100 x 50 mm rectangular steel columns and is modular to allow easy installation and removal.
Following tests in the closed-jet configuration, an open-jet configuration was evaluated using the same ground plane and force balance as the closed-jet configuration but without the wall and roof insert. This was done primarily to improve the operational efficiency of the tunnel and to minimize the downtime when converting from full-scale to reduced-scale configuration. Due to time and resource limitations, corrections were not applied to data from the open-jet test section at the time of this study. Based on the test section geometry and static pressure gradient profile, it is expected that the corrections would be small and therefore they could (mostly) be encompassed by the correlation factors. Further study is intended.
Tunel Aerodinamico 2 (TA-2) - Brazil
The TA-2 Wind tunnel was built during 1950-58 in the city of Sao Jose dos Campos, state of Sao Paulo, Brazil. Its initial role was to serve as a laboratory of aerodynamics for the course of aeronautical engineering at the Instituto Tecnologico de Aeronautica (ITA) and to be a basic tool for establishing and consolidating the aviation industry in Brazil, in particular the internationally known EMBRAER. The portfolio of tests includes cars, buses, ships, submarine, military artifacts, construction models, and various studies. However, the most extensive campaigns are those involving the aircraft of EMBRAER. Since the 1970s the TA-2 is a facility of the Instituto de Aeronautica e Espaco (IAE). The TA-2 is the largest wind tunnel in Latin America in commercial operation.
In 1968 load cells were installed and the data acquisition and reduction system became completely automated. In early 2000 the TA-2 underwent an update of its propulsion system relying on a 7-blade propeller of composite material and variable pitch. This reduced the time to turn the tunnel on and off and streamlined model configuration change processes. At that time, a PIV system (2D and S-PIV) and LDA were acquired, adding to the hot wire anemometer system (HWA / CTA) and the existing pressure sensors.
The tunnel has a closed-jet test section, and for automotive testing a ground board with an elliptical leading edge and dimensions of 3 m wide x 5 m long x 60 mm in thickness is installed at 275 mm above the floor of the tunnel (to the center of the nose fairing). An external pyramidal balance is situated below the wind tunnel test section floor and capable of measuring six components; it also allow tests in yaw conditions. The balance connects to the model through a shrouded mast located 2.37 m from the ground board leading edge. At the top of the mast there is an H-shaped model support structure within the ground board. This H-shaped support has four adjustable pins that support the model through its tires.
The dynamic pressure of the tunnel is calibrated against a pitot-static probe at the model's position to provide a reference for the aerodynamic coefficients.
A wind tunnel correlation project should ideally comprehend the major test model variables including basic geometry, size, attached vs. separated flow, wake structure, internal and underbody flows, test conditions, and the associated ranges of forces and moments expected during subsequent test programs. Reliably repeatable configurations and procedures from one test to the next are essential. Due to practical limitations and available resources, it is seldom possible to conduct an ideal study.
This correlation project utilized a 1/3-scale model with interchangeable rear modules that provided four aerodynamically significant geometries: notchback sedan, fastback, squareback (i.e. station wagon), and pickup truck (comparable to an Australian "ute"). Add-on parts including a front air dam, grille covers, rear spoiler, and wheel covers provided additional configuration variants. The model was constructed completely of rapid prototype (SLA) parts on an aluminum structure, and had a smooth, painted surface. This provided an overall shape representative of production vehicles with a high level of realistic detail. The model was based on a Buick Verano, although details of its construction and differences such as a less-restrictive engine compartment flow path produced a baseline drag coefficient ([C.sub.D]) greater than that of the production vehicle. This was not an issue for correlation, as the primary interest was repeatability from test to test. Model dimensions are listed in Table 7. The model and the four rear end configurations are shown in Figures 13, 14, 15, 16, 17.
The test conditions for most runs were based on the standard GMAL test speed of approximately 195 km/h, at 0[degrees] yaw angle. Reynolds number sweeps and yaw sweeps were performed with each rear end shape. A continuity-based blockage correction was applied to the data from the closed-jet wind tunnels. Although this is a somewhat simplistic approach, it has been demonstrated to work reasonably well for small blockage ratios . No corrections were applied to the data from the Monash open-jet tunnel, as previously discussed. No other corrections based on individual tunnel flow characteristics were applied. For example, while the static pressure gradients produced a horizontal buoyancy force, due to the small size of the model and the reasonably flat gradients, the largest buoyancy offset was approximately [DELTA][C.sub.D] = 0.002. Thus it was considered acceptable to simply include the differences in buoyancy in the statistical correlation.
Using a single model with multiple geometries greatly simplified shipping and logistics, while still providing a wide range of configurations and associated forces and moments. Test time was reduced because only a single model had to be installed and set up on the balance. The model was relatively light in weight and could be easily handled by two people. Interchanging parts during model configuration changes was simple and quick.
The rear end modules produced significantly different flow fields ranging from well-attached flow with a reasonably small wake, to highly separated flow with a large, unsteady wake. Blocking the front inlets provided a reliable step change in drag and lift levels due to the change in flow split over, under, and through the model. The construction of the model from hard parts provided good repeatability from test to test. Its durability was proven throughout its global travels.
To best ensure consistency among multiple tests, it is beneficial to have at least one person attend and oversee all test events. This was not possible during this study, but it was intended that all model configurations would be easily repeated and well documented. Written procedures, photographs, and real-time consultation were valuable in this regard.
It was observed during the course of the project that some parts, such as the air dam and wheel covers, were not positively located and mechanically attached. As a result, they were not always installed identically and therefore produced variations in the results. Those data points were not included in the final analysis.
Parts made from SLA materials may deform over time and are susceptible to moisture absorption. This concern was addressed through repeated tests at GMAL during the three year duration of the project, before and after shipments to other tunnels, and even after the model in its crate was found floating in food waters. (Fortunately this occurred after the completion of the correlation tests!) As an example, for the eight configurations (four rear ends with open and closed grilles) that were tested after the food, the average change in [C.sub.D] at 0[degrees] yaw vs. the original test was 0.0003.
While the single, modular model was an important enabler for the completion of this study, the scope of the correlation was limited by having only one size of test model. This did not allow evaluation of the effectiveness of the blockage corrections or of potential interactions between larger models and the individual test section configurations. When possible during test events subsequent to the correlation study, it has been helpful to conduct comparison runs at two tunnels to evaluate the behavior of those models relative to the established correlation.
The discussion of data herein is limited to those components of forces and moments that are significant at 0[degrees] yaw angle. This does not suggest that aerodynamic influences at non-zero yaw are unimportant, but the priority of fuel economy improvement tends to focus aerodynamic development on drag reduction with a secondary consideration of lift and pitch moment as they influence vehicle handling. While data was collected at non-zero yaw angles, the sample size is somewhat limited and analysis of that data is beyond the scope of this paper.
In Figures 24, 25, 26, 27, 28 in the appendix, the coefficients of drag, lift, front lift, rear lift, and pitching moment measured in each wind tunnel are plotted against the averages of the values from all tunnels. The results are sufficiently consistent and linear to enable the definition of first-order correlation equations, which may be used to convert test results from one wind tunnel to equivalent values as though measured at another tunnel. This permits comparisons and tracking of the status of future vehicle development programs regardless of the source of the data. For convenience, the polynomial coefficients of the correlation equations are listed in Tables 8, 9, 10, 11, 12.
As might be expected, the spread in lift data was larger than for drag. This is a typical result because lift force is the result of pressures integrated over the large projected plan area compared to drag, which is related to frontal area. However, it is interesting that there was a large spread in front lift, while rear lift was more tightly grouped. Considering differences in wind tunnel configuration, the tunnels with boundary layer suction produced higher than average front lift, while those with elevated ground boards produced lower values. This suggests the presence of upwash near the front of the model, possibly due to insufficient suction and flow over the lip of the suction scoop. Alternatively, the acceleration around the nose of the elevated ground board may have produced a local reduction in static pressure under the front of the model, reducing front lift. A detailed mapping of the flow would be required to confirm these theories.
Because aerodynamic development testing is an iterative process, it is important that wind tunnels produce consistent increments between model configurations. Figures 18, 19, 20, 21, 22, 23 demonstrate that the increments were consistent in direction and comparable in magnitude among all of the wind tunnels. In addition, the differences in increments were reduced after applying the correlation equations to normalize the results to equivalent "average tunnel" values. For the sake of space, only the "Ute" increments are shown.
CONCLUSIONS AND RECOMMENDATIONS
While all wind tunnels are simulation tools with differences and compromises in their ability to replicate the real world, the results of this correlation study demonstrate that the five wind tunnels and six tunnel configurations produce sufficiently similar results to be used interchangeably for reduced-scale aerodynamic development. The test data revealed consistent trends among the tunnels, enabling the definition of correlation equations that permit application of test results regardless of the source. Comparable increments between model configurations mean that development testing in any of the tunnels will drive the same design decisions.
The single model with interchangeable rear body modules facilitated an efficient test process and simplified the logistics of testing on four continents. The use of SLA rapid prototype components provided a representative vehicle shape with a high level of detail, in a lightweight model that was easy to handle. Although there are concerns about the stability of the material, repeated tests demonstrated that this was not an issue throughout the duration of the project.
Experience with this project provided several lessons and recommendations for successful correlation projects:
* A modular model with aerodynamically significant shapes permits efficient collection of a large set of data.
* Models of more than one size (especially frontal area) will produce a more comprehensive correlation with improved application to a range of future test objects.
* The model must be durable and stable, and parts must be positively attached in a manner that is easily and accurately repeated.
* The test procedures and configurations must be extensively documented with instructions, dimensions, and photographs. It is strongly recommended to have at least one person who participates in all tests to ensure consistency.
[1.] Barlow, J., Rae, W., and Pope, A., "Low-Speed Wind Tunnel Testing," (New York, John Wiley & Sons, 1999), 19-22, ISBN 0-471-55774-9.
[2.] Kelly, K., Provencher, L., and Schenkel, F., "The General Motors Engineering Staff Aerodynamics Laboratory--A Full-Scale Automotive Wind Tunnel," SAE Technical Paper 820371, 1982, doi:10.4271/820371.
[3.] University of Maryland College Park, "About the Glenn L. Martin Wind Tunnel," http://windvane.umd.edu/about/index.html, accessed Sept. 30, 2015.
[4.] Arnette, S. and Sung, B., "Aerodynamic Commissioning for the Korea Aerospace Research Institute Low Speed Wind Tunnel", AIAA-2000-0291, 38th Aerospace Science Meeting, 2000.
[5.] Yang, Z. and Schenkel, M., "Assessment of Closed-Wall Wind Tunnel Blockage using CFD," SAE Technical Paper 2004-01-0672, 2004, doi:10.4271/2004-01-0672.
[6.] SAE International Surface Vehicle Recommended Practice, "Vehicle Aerodynamics Terminology," SAE Standard J1594, Rev. July 2010.
The authors would like to acknowledge the contributions of Donnell Johnson and Robert Schaffer of General Motors, for their roles in the design and construction of the correlation model and support of the test program. In addition, the successful completion of the project was due to the efforts of many people at the individual wind tunnels.
[C.sub.D] - Drag coefficient
[C.sub.L] - Lift coefficient
[C.sub.LF] - Front lift coefficient
[C.sub.LR] - Rear lift coefficient
[C.sub.PM] - Pitching moment coefficient
Correlation graphs are presented in this appendix in single-column format for legibility.
General Motors Co.
GM Holden Ltd.
GM do Brasil
GM Korea Co.
University of Maryland
Korea Aerospace Research Institute
Luis Fernando Gouveia de Moraes
Institute of Aeronautics and Space
Table 1. GMAL Specifications Year opened 1980 Test section type Closed jet, stationary ground Nozzle/test section area 56.3 [m.sup.2] Nozzle/test section H x W 5.44 x 10.36 m Contraction Ratio 5:1 Maximum speed / speed for flow specs 62 m/s / 55 m/s Boundary layer control Suction scoop Boundary layer displacement thickness [delta]* = 5.4mm@X = -0.8m Static pressure gradient near model dCP/dX = -0.0017/m Turbulence Intensity 0.5% Corrections Continuity Table 2. GLMWT Specifications Year opened 1949 Test section type Closed jet, stationary ground Nozzle/test section area 7.7 [m.sup.2] Nozzle/test section H x W 2.30 x 3.36 m (with fillets) Contraction Ratio 7.5:1 Maximum speed / speed for flow specs 103 m/s / 55 m/s Boundary layer control Suction scoop Boundary layer displacement thickness 5* = 3.0mm@X = -0.8m Static pressure gradient near model dCP/dX = 0.0018/m Turbulence Intensity 0.21% Corrections Continuity Table 3. KARI LSWT Specifications Year opened 1998 Test section type Closed jet, stationary ground Nozzle/test section area 12.0 [m.sup.2]/8.4 [m.sup.2] (total/above groundplane) Test section H x W above groundplane 2.10 x4.00 m Contraction Ratio 8.24:1 Maximum speed / speed for flow specs 120m/s/55m/s Boundary layer control Elevated ground board Boundary layer displacement thickness [delta]* = 3.48mm@X = -0.8m Static pressure gradient near model dCP/dX < 0.003/m Turbulence Intensity 0.13% Corrections Continuity Table 4. Monash Closed-Jet Specifications Year opened 2011 (closed-jet section upgrade Test section type Closed jet, stationary ground Nozzle/test section area 10.40 [m.sup.2]/7.92 [m.sup.2] (total/above groundplane) Test section H x W above groundplane 1.98 x 4.00 m Contraction Ratio 5:1 Maximum speed / speed for flow specs 56 m/s / 48 m/s Boundary layer control Elevated ground board Boundary layer displacement thickness [delta]*=4.9mm@X = -0.8m Static pressure gradient near model dCP/dX = -0.003/m Turbulence Intensity 1.4% Corrections Continuity Table 5. Monash Open-Jet Specifications Year opened 2013 (open-jet upgrade) Test section type 3/4-open jet, stationary ground Nozzle/test section area (total/above groundplane) 10.40 [m.sup.2]/7.92 [m.sup.2] Nozzle H x W above groundplane 1.98 x 4.00 m Contraction Ratio 5:1 Maximum speed / speed for flow specs 56 m/s / 48 m/s Boundary layer control Elevated ground board Boundary layer displacement thickness [delta]*=4.9mm@X = -0.8m Static pressure gradient near model dCP/dX - 0.005/m Turbulence Intensity 1.4% Corrections None Table 6. IAE TA-2 Specifications- Year opened 1950 Test section type Closed jet, stationary ground Nozzle/test section area 6.30 [m.sup.2]/5.39 [m.sup.2] (total/above groundplane) Test section H x W above groundplane 1.80 x 3.00 m Contraction Ratio 14.6:1 Maximum speed / speed for flow specs 125 m/s / 56 m/s Boundary layer control Elevated ground board Boundary layer displacement thickness 5[delta] = 3.77mm@X = -0.8m Static pressure gradient near model dCP/dX - -0.0065/m Turbulence Intensity 0.7% Corrections Continuity Table 7. Correlation model dimensions Model Full Scale Frontal area ([m.sup.2]) 0.2477 2.229 Wheelbase (mm) 895.7 2687 Track - front/rear (mm) 512.0/514.0 1536/1542 Overall length (mm) 1555.0 4665 Overall width (mm) 598.0 1794 Overall height (mm) 490.0 1470 Table 8. Correlation coefficients for measured [C.sub.D] vs. the average Wind Tunnel Slope (CD) Intercept (Cd) GMAL 1.0064 +0.0003 GLMWT 0.9773 -0.0006 KARI LSWT 0.9645 +0.0029 Monash Closed Jet 1.0096 -0.0037 Monash Open Jet 0.9919 +0.0090 TA-2 1.0447 -0.0055 Table 9. Correlation coefficients for measured [C.sub.L] vs. the average Wind Tunnel Slope (Cl) Intercept (Cl) GMAL 0.9975 +0.0109 GLMWT 0.9268 +0.0178 KARI LSWT 0.9499 -0.0040 Monash Closed Jet 1.0487 -0.0152 Monash Open Jet 1.0398 -0.0111 TA-2 1.0347 +0.0014 Table 10. Correlation coefficients for measured [C.sub.LF] vs. the average Wind Tunnel Slope (Clf) Intercept (Clf) GMAL 1.0469 +0.0102 GLMWT 0.9748 +0.0239 KARI LSWT 0.9860 -0.0100 Monash Closed Jet 1.0196 -0.0093 Monash Open Jet 1.0769 -0.0105 TA-2 0.8899 -0.0045 Table 11. Correlation coefficients for measured [C.sub.LR] vs. the average Wind Tunnel Slope (Clr) Intercept (Clr) GMAL 0.9875 +0.0003 GLMWT 0.9446 -0.0078 KARI LSWT 0.9676 +0.0044 Monash Closed Jet 1.0346 -0.0048 Monash Open Jet 1.0265 -0.0003 TA-2 1.0391 +0.0082 Table 12. Correlation coefficients for measured [C.sub.PM] vs. the average Wind Tunnel Slope (Cpm) Intercept (Cpm) GMAL 1.0117 +0.0057 GLMWT 0.9692 +0.0165 KARI LSWT 0.9865 -0.0068 Monash Closed Jet 1.0185 -0.0026 Monash Open Jet 1.0419 -0.0044 TA-2 0.9696 -0.0087
|Printer friendly Cite/link Email Feedback|
|Author:||Meinert, Frank; Johannessen, Kristian; Saito, Fernando; Song, Bongha; Barlow, Jewel; Burton, David;|
|Publication:||SAE International Journal of Passenger Cars - Mechanical Systems|
|Date:||Jun 1, 2016|
|Previous Article:||A numerical and experimental evaluation of open jet wind tunnel interferences using the DrivAer reference model.|
|Next Article:||Adjoint-driven aerodynamic shape optimization based on a combination of steady state and transient flow solutions.|