Wind tunnel modified for evaluation of heavy-vehicle aerodynamic drag

  • 30-Oct-2013 04:24 EDT

1:3 scale model in the Monash University 1.4-MW wind tunnel.

Significant progress has already been made in developing solutions that reduce the aerodynamic drag of heavy vehicles; however, there still remains plenty of work to be done to increase the uptake of these proven aerodynamic devices such as boat-tails, gap seals, and side skirts. Active flow control devices are immature and remain at a level where they are not being implemented in on-road situations.

A challenge to the uptake of drag-reduction devices on the road has been the lack of large-scale wind tunnel testing at representative Reynolds numbers. The majority of experimental development of heavy-vehicle aerodynamic technology has been completed at a scale of around 1:10, often on simplified bodies. At such a small scale, it becomes difficult to accurately represent detailed geometry, and very high wind speeds are required to replicate on-road conditions; this can push the power limits of many wind tunnels and lead to issues with compressibility.

In an attempt to achieve a balance between Reynolds number, model scale, and tunnel blockage issues, a 1:3 scale heavy-vehicle testing program has been initiated at the Monash University 1.4-MW wind tunnel.

The Monash University 1.4-MW wind tunnel is a closed circuit type tunnel with a 3/4 open jet automotive test section. The tunnel is powered by two 0.7-MW fans. Between the fans and the jet exit, the flow is contracted twice, first by a two-sided horizontal contraction and then a one-sided vertical contraction.

In standard configuration, the jet is 4 m (13 ft) wide and 2.6 m (8.5 ft) high giving a jet area of 10.4 m2 (112 ft2) and a total contraction ratio of 5:1. The test section is 9 m (30 ft) long from the jet exit to the beginning of the collector. The origin of the coordinate system for the tunnel is the center of the turntable, with the Z axis pointing vertically and the X axis in the downstream direction. Previous testing on a 1:3 scale GTS model showed that this configuration was unsuitable for testing a model this large as the shear layer emanating off the roof of the jet impinges on the shear layer at the back of the truck.

A number of modifications were made to the tunnel to improve test conditions for heavy-vehicle scale testing. The angle of the vertical contraction was reduced, raising the height of the jet exit to 3 m (10 ft), combined with a 0.27-m (0.89-ft)-high false floor for a total jet area of 10.9 m2 (117 ft2) and a contraction ratio of 4.8:1. This had the effect of raising the shear layer off the top of the jet out of the vicinity of the model as well as reducing the solid blockage from 11.2% to 10.6%. This comes at the expense of a slightly reduced top speed, increased free stream turbulence, and decrease in flow uniformity.

As well as being raised, the jet exit was moved upstream by 0.9 m (3 ft). This increased the spacing between the nose of the model and the jet exit, minimizing nozzle blockage effects. To reduce negative pitch in the flow, the vertical contraction was moved a further 2 m (7 ft) upstream, with the final section of the jet being of constant cross section.

Force and moment are determined using four Kistler type 3-component force transducers mounted in the wind tunnel floor. Supports extend down between the trailer wheels and are mounted to the force balance beneath the false floor. The force balance is designed such that three components of force (drag, side, and lift forces) and three moments (pitch, roll, and yaw) are determined. The force balance is mounted to the turntable so that at yaw angles the forces are measured in the model axis and no transformation is necessary.

Flow field measurements were taken with a four-hole dynamic pressure probe (cobra type). The cobra probe is capable of obtaining three components of velocity, static pressure, pitch, yaw, turbulence intensities, and Reynolds stresses. The probe can only make accurate readings for flow at incidence angles of less than 45° and is unsuitable for measuring reversed flow. The precision of the velocity outputs from the cobra probe is 0.1 m/s. Flow mapping was done at a test section velocity of 100 km/h (62 mph), corresponding to the maximum rated velocity for the probe traverse. Samples were taken at a rate of 500 Hz for 15 s.

All pressure measurements, including those from pitot static tubes placed in the test section, were taken using a dynamic pressure measurement system (DPMS) from Turbulent Flow Instrumentation. The DPMS has 64 available diaphragm channels, all connected to a common reference that was placed in a plenum outside the flow of the test section. The maximum range of the system is ±3.0 kPa. A transfer function is incorporated into the data-acquisition system to remove the effect on the transient response due to the length and diameter of the pressure tapping tube used.

To assess the flow quality in the modified test section, measurements were taken of flow uniformity, boundary and shear layer profiles, and streamwise static pressure gradient. Measurements of the side shear layer with a long vehicle model at yaw were made, and it was determined that the presence of the model deflected the shear layer but did not change its shape.

The modified test section has a longitudinal static pressure gradient of 0.0021 m-1, which is quite high, so aerodynamic drag results need to be corrected for this. In addition, care must be taken when designing diffusing surfaces such as boat-tails, especially when in regions of positive static pressure gradient.

The boundary layer has a displacement thickness δ* = 12 mm at the front of the model, and the tunnel is capable of a top speed of 47 ms-1, which is equivalent to a width-based Reynolds number of 2.4 x 106.

This article is based on SAE International technical paper 2013-01-2455 by Damien J. McArthur, David Burton, Mark Thompson, and John Sheridan of Monash University.

HTML for Linking to Page
Page URL
Rate It
5.00 Avg. Rating

Read More Articles On

Exa and Navistar researchers examine CFD simulations and wind-tunnel test results to increase the understanding of simulation methods used to reveal aerodynamic losses in the real-world environment.
Many among the 13 partners that jointly developed the Genius CAB concept are firmly established in the automotive industry, and quite of few of these suppliers note the influence their automotive arms played in their contributions to the project.
GH100 and GH101 hoses from Eaton are suited for high-percentage biodiesel blend and high-temperature oil applications.
Active flow control devices based on electric plasmas can smooth the turbulent wakes that sweep off the rear edges of truck trailers travelling at highway speeds.

Related Items

Technical Paper / Journal Article
Training / Education
Technical Paper / Journal Article
Technical Paper / Journal Article
Technical Paper / Journal Article
Technical Paper / Journal Article
Technical Paper / Journal Article
Technical Paper / Journal Article