Angle of the contraction cone in an open circuit wind tunnel

[deleted-113375]

Hello everybody

My name is Alen and I am currently building a wind tunnel as my complementary work in school. For most of my plannings and designs I considered the book "Low-Speed Wind Tunnel Testing".
During the plannings of the contraction cone, I encountered a blur: It is written that the area ratio ought to be in the range of 7-12.
But what about the lenght (and its resulting angle of course) of the contraction cone? Is there an effect on wind-speed or flow similarity or can I just neglect the lenght/angle of the contraction cone?

Thank you very much for your help!


Best wishes

Alen

[deleted-93346]

I know almost nothing about wind tunnels. I am seeking help from our other experts, but your question is so specialized that I'm not sure anyone can help. If you can describe the problem well enough, someone may know enough about low speed aerodynamics to give an answer.
Meanwhile take a look at this forum:

http://www.eng-tips.com/threadminder.cfm?pid=1

They may do better than we can to help out. Best of luck.

[theborg]

Alen94,
Welcome to Science Buddies and thank you for your question. To start, the geometry of the contraction zone of a wind tunnel is very important. As the input into the test section, the contraction zone must ensure a quality flow. The fluid dynamics involved in optimizing the conditions are very advanced. That being said, if you want to do some research into a couple of key components of uniform flow in a low speed wind tunnel, look-up the following terms:

1) Reynolds Number
2) Boundary Layer

But, a good design can be reached without characterizing all the parameters to the greatest detail. A good rule of thumb for the length of the contraction zone is at least as long as the mouth is wide/tall. However, for a ratio of 7, this can end up being quite large and long. After all, space available must be considered as well. Science Buddies actually has a good design that can be built at home with materials readily available. It should be adequate for any testing you would want to do in school. The design can be found at the following link:

https://www.sciencebuddies.org/science- ... -toc.shtml

If you still have questions, please post back.

[deleted-113375]

Good Day and thank you very much for your answers !

@ John Dreher: Thank you for your hint to contact the engineering forum. I already posted the question and I quickly received some answers. They told me that the contraction angle shoud be around 15 degrees to ensure flow similarity. Now I asked them how to reason this number. Does anybody of you have an idea?

@ theborg: I already heared about his rule of thumb. But what is the reason for these measures? I had a look at your wind tunnel (sciencebuddies) and I saw that you didn't follow this rule. What about the results? Was the flow laminar? I think your nozzle has a contraction angle of around 20 degrees. I'm thinking of building one with the same angle.

Additionally I will post an extract of my 3D-model, just to be sure you understand what I mean:



Uploaded with ImageShack.us


Best wishes

Alen

[theborg]

Alen94,
Just to clarify, the tunnel on the site is not my design. I haven’t worked with this particular tunnel design. However, I’ve worked with quite a few low and high speed tunnels before…one of which was very similar to the one depicted in Figure E on the first page of the Science Buddies design.

Both the 15 degrees, noted by the engineering forum and the one I pointed out are “rules of thumb”. That being said, that means they are ballpark figures designed to give decent results without an intensive flow analysis that would be required for a professional wind tunnel (very expensive, and something not done by hand).
Ideally you’d like to have the angle as small as possible. The “rule of thumb” (i.e. length approximately equal to inlet height) I mentioned earlier gets you close to a 15 degree angle (~17 deg). However, for the design I mentioned, the height and length of the contraction cone is constrained by standard materials available at most hardware stores. Sheets of wood typically come in 2’ X 4’ rectangles. For an area ratio of 7 and a test section opening of 12x12 inches, the inlet of the tunnel must be ~32 in. One can make a side out of a single sheet of commercially available wood if you “cheat” a bit on the contraction cone length and angle. Again, for slow speed flows, this is unlikely to be an issue.

In order to get closer to a 15 degree turn, you’d have to find a non-standard size piece of wood, or make each wall of the cone out of two pieces, which would create a seam somewhere in the middle of the cone that could “trip” your boundary layer flow prior to entering your test section, causing unpredictable turbulent regions in your upstream flow (Ux). This is very undesirable, much better to make your flow turn an extra 5-7 degrees to get into the test section. Think of a skate boarder rolling down a sidewalk, then suddenly there is a slight lip where the sections of two pieces of the sidewalk don’t match up. If this lip is significant enough, the skater is unlikely to remain on his board.

Ok, bottom line: If you are looking for a simple equation that tells you the optimum geometry, there really isn’t one…There is a paper by Bell and Mehta (April 1988) that has a good description of the considerations required to optimize the design of this component of a wind tunnel, at the following link:

https://moodle.polymtl.ca/file.php/1047 ... l_1988.pdf

[deleted-71588]

Hello everybody
Is there an effect on wind-speed or flow similarity or can I just neglect the lenght/angle of the contraction cone?

The shape and length of all cross-sectional areas in duct work affect back pressure and similarity of flow. With enough knowledge of fluid dynamics and computer programming, one might be able to model an arbitrary system and compute its characteristics. To do that, one would have to assume some input fluid flow. Unfortunately, there is a huge effect on any fan used to create this input fluid flow and the assumptions won't hold.

Long before there was a computer system capable of modeling fluid flows in reasomable time at reasonable expense, people were designing and experimenting with duct work for forced air heating and cooling and wind tunnels.

From experimentation, people found by measuring pressures and flows in square and round cross sectional areas that most turbulence, air pressure differences, and air speed was limited to the first 3 diameters (or inscribed diameters) worth of length.

They told me that the contraction angle shoud be around 15 degrees to ensure flow similarity. Now I asked them how to reason this number. Does anybody of you have an idea?

The design you pictured is symetric, so if each angle measures 15 degrees, you actually have 30 degress of narrowing. Trigonmetry Tangent of an angle is the ratio of the opposite (diameter) over the adajacent (length of transition) side of a right triangle. If you want the transition to be at least 3 times the final diameter, tan x = 1/3 results in x = 18 degrees.

But lets look at your fan source. Are you planning on using a simple 30" square box window fan? If yes, then the motor area in the middle is not moving air and the four corners are not moving air. You are just going to have to build one and measure the pressues and speeds and see how even they turn out. If the flow isn't laminar enough, you can always lengthen the straight section.