Existing Wind Tunnels


NWT Model

Turbulence Research Laboratory


Turbulence has been studied for a long time, but is nevertheless often cited as the last unsolved problem in classical mechanics. Many classical ideas of turbulence theory date back to the 1930s and 40s. These ideas have evolved since, but it has never been possible to truly test most of them because of the absence of large and long enough, high quality (low background turbulence) research facilities.

Experimental work in a large research wind tunnel is needed to solve a number of fundamental questions and enable scientists and engineers to further improve computations of turbulent flow. Simply put, the problem is how to achieve a separation of length scales, energetic to dissipation, of 10^5 or larger – and still be able to resolve the smallest scales occurring in the flow with the smallest technically feasible probes (approximately 10 x10^(-6)m).

Existing research wind tunnels are too small to reach high enough Reynolds numbers while still permitting resolved measurements of the smallest scales. They are either too short for the turbulence to evolve from its upstream (initial) conditions, too narrow for the large energetic turbulence scales to be free from the influence of the walls, or have too high a background disturbance (free-stream turbulence) level to extract the features of primary interest.

Despite increases in computational power and progress in numerical techniques, it is currently not possible to resolve the small scales at high Reynolds number. Even with computational power doubling every 18 months, it would take several decades before a model-free, direct numerical simulation (DNS) of the simplest flow case of non-decaying isotropic turbulence could be performed with the separation of scales 10^5 equivalent to the wind tunnel proposed here (resolution requirements for a comparable simulation: number of grid points > 10^15).

The Nordic Wind Tunnel overcomes the shortcomings of present research facilities and is proposed for construction at Chalmers University of Technology. It would be
• wide enough to remove the effect of side walls on the energetic turbulence scales
• fast enough and large enough to get the necessary high Reynolds numbers, yet still resolve the dissipative scales
• long enough and with low enough background disturbances to obtain the necessary downstream development times and will thus provide an experimental facility capable of resolving some of the oldest questions in turbulence while also testing conclusively new ideas.

In the unique inverse design process it was asked what length and time scales needed to be resolved to conduct “meaningful” measurements (e.g., resolution of wall layer to obtain shear stress) and what Reynolds numbers needed to be achieved in the experiments to be performed (e.g., boundary layers, far wake, decay of isotropic turbulence) to help resolve fundamental questions and sort competing theories. These length and time scales and Reynolds number criteria depend on the flow being measured. The size (length and cross-sectional area of the test section) and performance (maximum free stream velocity) of the proposed wind tunnel facility were then determined by what can be resolved with existing probes.

Example of one design criterion:
For wall-bounded turbulent flows the requirement was that one viscous length scale can still be resolved (using a micro-LDA or a micro-PIV system, with a measuring volume height of 10 x10^(-6)m) while achieving a Reynolds number based on momentum thickness of at least 100,000.
Other “base case flows”, e.g. decaying turbulence or wake flows, also provided further size and flow quality criteria.

Result of the inverse design/dimensioning process is a wind tunnel with a
- test section of 40m length,
- cross section after contraction 3x3m,
- maximum free stream speed 40m/s and a
- free stream turbulence u’/U of < 0.02%.
The total dimensions of the facility are: length: 79m, width: 21m, height 7m.

Scale model of the Nordic Wind Tunnel: The small scales of the flow become smaller with increasing flow speed or pressure. To quote the late Professor Tony Perry, University of Melbourne, who realized the dilemma and stated that: “Big and slow - is the way to go”.


created by MW