[BLANK_AUDIO] Welcome back to Sports and Building Aerodynamics in the week on wind-tunnel testing. In the seventh module we're going to focus on best practice guidelines. And we start again with the module question. The blockage ratio is defined as the ratio between the frontal area of the wind-tunnel and the area of the wind-tunnel cross-section. And it should be sufficiently low to avoid artificial flow acceleration around the model, due to the presence of the wind-tunnel walls, which of course, are not presence there in reality. How large can or should the blockage ratio be, without having to apply corrections? Is that A, lower than 1%, B, lower than 5%. C, lower than 20% or D, lower than 50%. Hang on to your answer and we'll come back to this question later in this module. At the end of this module, you will understand some important aspects of best practice in wind-tunnel testing, in particular, as related to wind engineering applications. You'll understand the blockage ratio and its importance, and the importance of consistent modeling of all length scales. You'll also see some main issues involving Reynolds number scaling. In this module, we're only going to give a very brief view on best practice and this is indeed, as mentioned before, specifically focused on wind engineering. There's much more extensive information that can be found in many excellent textbooks, and these are a few of those. Concerning wind-tunnel simulation of the ABL, we've already stressed a few times this week that for many wind engineering applications, we focus on strong wind, and therefore the flow is not very much influenced by the thermal stratification. So, we focus on isothermal boundary layers. Then typical ABL wind-tunnel dimensions are the following. The test section width can range from 2 up to 5 meter. The length is typically 15 to 30 meter. We use air at atmospheric pressure, and the operating speed is in the range of 10 to 50 meter per second. Sometimes shorter tunnels are used, and these tunnels then have to be equipped with spires, barriers, and so on, so additional features to generate the appropriate roughness profiles. And the wind tunnel shown here is not a wind tunnel that is too short. But it is a wind tunnel from which we got this very nice photograph, clearly illustrating the different roughness features. And the tests are generally performed for so-called stationary approach-flow properties. This means that we have a constant mean wind speed and constant RMS values for turbulence. The approach wind, as stated by ASCE, should actually be simulated to a large degree and to the maximum degree of reproducibility of natural wind conditions over different types of full-scale terrain. This means indeed that, depending on the type of terrain roughness, we should apply different combinations of the roughness features. The minimum modeling requirements are, that we need to reproduce the mean wind speed profile, and also the intensity of the longitudinal turbulence component, that is shown here on the right side in this picture. Then also the important properties of atmospheric turbulence should be reproduced. Certainly the relevant length scales, especially of the longitudinal turbulence component. And then finally we should make sure that we have a zero, or near-zero longitudinal pressure gradient as in reality. Concerning topographic models, sometimes in case of complex terrain, it is important that we make a first wind-tunnel run with small-scale topographic models to get more information about the approach-flow conditions, and then later on this information can be used for modeling wind at larger scales and finally, getting accurate results. Concerning nearby buildings and structures: they often influence the approach flow to a very large degree, so they must be included in the model. And often it's stated that typically all major buildings and structures within about 300 to 800 meter from the site of interest, have to be included in the scale model. And that then often means, that this model will cover the entire wind-tunnel turntable or even more than the wind-tunnel turntable, as shown in this photograph. But, it is true that as you move away from the building of interest, or the site of interest, that the degree of model detail that has to be included, can be decreased. So for example, balconies and so on, and other facade and roof details from far away buildings, do not need to be included in the testing. Then concerning geometric scale. Well there are different ratios that we can focus on. And three of those are given here. Where this first one is the Jensen number, so it's the ratio of the building dimension to the aerodynamic roughness length. Then we have the second one which is building dimension to gradient height, and the last one is building dimension to the turbulence length scale. And depending on the type of test, one of these will sometimes be given larger priority than others. For example, for low-rise buildings the aerodynamic roughness length can be expected to be more important than the gradient height. So that's why, the first criterion will then be focused on, and then the second one might be relaxed. And for bridges for example, where we sometimes do not include the ground surface, well then the first criterion might be relaxed. Typical geometrical scales used in wind-tunnel tests for large buildings and structures are from 1:300 or 1:600. But for small buildings and structures, and we want to focus then on the small-scale turbulence effect, then often scales are used of 1:100 or larger 1:50, 1:40. Then turning to the blockage ratio. Well as mentioned before, it's the ratio, of the frontal area of the model to the cross-section of the wind-tunnel, and when this ratio is too large, we'll get artificial acceleration, we get a speed up of the flow, so the blockage ratio preferably should be lower than 5%. If it is larger than 5%, then certainly corrections should be applied. Concerning Reynolds number scaling, we want to repeat again what was mentioned in previous modules, that generally it is not possible. But this is also not a major disaster, because when we have buildings with sharp edges, separation will take place there. And the Reynolds number independence can fairly easily be achieved although we do not have Reynolds number matching. Sometimes indeed also for rounded shapes a roughening of surfaces is applied, but we try to have a Reynolds number larger than 10,000. This is a typical threshold that is used. If you expect consequences of Reynolds number scaling, then it is important to evaluate them through comparison with either full-scale data, or doing selected tests with a larger scale model, or just doing tests for the same model, but over a wider range of wind speeds. So coming back to the module question. How large should a blockage ratio be, or can it be without having to apply corrections? The answer is B, lower than 5%. In this module, we've learned about some important aspects of best practice in wind-tunnel testing, particular as related to wind engineering applications. We've focused on the blockage ratio and its importance. On the importance of consistent modeling of all length scales and on some issues of Reynolds number scaling. This concludes our week on wind-tunnel testing. In this week, we have learned about the importance of wind-tunnel testing, the different types of wind tunnels, the difference between aeronautical wind tunnels on the one hand, and atmospheric boundary layer wind tunnels on the other. The main components of such wind tunnels. And the importance of flow quality and similarity. I hope you have enjoyed this week on wind tunnel testing. In the next week, we're going to focus on Computational Fluid Dynamics. Thank you again very much for watching. And we hope to see you again in the next week. [BLANK_AUDIO]