Low Speed Wind Tunnel questions

[bikeboy]

I am building a small wind tunnel, and I have a few questions about it.

First: The Fan I am using produces 800cfm and is 12'' in diameter, I am going to narrow it down to 6''. The calculated wind speed that I got was 74.5km/h. Can anyone back me up on this?

Second: When I use scale models of say wing foils or other things of that nature should the wind speed in the tunnel be higher or lower then in a full scale test. And can anyone point me in the direction of any equations to find these values?

Third: What is a cost effective smoke device?

-Adam

 

[FredGarvin]

1) Looks about right.

2) Start looking into fluid mechanics and the Buckingham Pi Theorem and dimensional analysis. That is how test parameters are selected for scaled models.

3) It depends on what "cost effective" means.

 

[desA]

You need to stabilise the airflow after the fan through a series of screens, or even balls. To accelerate the flow without stabilising will cause many tears in the test section. Go bigger after the fan, before converging to the test section.

For the fan, the more blades you have, the better - in terms of blade-passing frequency.

Rear, exit control & pressure-stabilisation is as important as front-end 'noise' control.

desA

 

[russ_watters]

2. Fred, why can't he just scale the Reynolds' number?

3. A cost effective smoke generator would be an incense stick...

 

[minger]

elaborating on #2 a little, basically what people do in practice is what's called non-dimensionalizing their equations and parameters. It would be easy to just scale down the size and then the speed. However, we find that certain ratios or non-dimensional numbers must remain the same.

I would say go with Fred and get a good understanding of dimensional analysis, because when you run your test, you're going to get a lot of numbers that won't really make any sense until you do.

 

[FredGarvin]

Russ, you may be right that at low enough speeds, the Re is all they would have to worry about. I'd have to look at some old notes on that one. I seem to remember more than Re being a concern.

 

[RainmanAero]

Hi Fred,

Russ, you may be right that at low enough speeds, the Re is all they would have to worry about. I'd have to look at some old notes on that one. I seem to remember more than Re being a concern.

You are right, but only as Mach numbers approach transonic. For low speeds (< Mach 0.6) it is sufficient to match the Reynolds Numbers. As one moves above this rule-of-thumb threshold it becomes more and more important to match Re and Mach. At least that is what I've been teaching them for 15 years now! All of this assumes steady, uniform flow. Once you introduce vorticity into the flowfield you've got bigger problems!

Rainman

 

[bikeboy]

Thanks, everyone. I think I got it, if you still want a little more explanation for 2 what I was saying is is I want to test a say 1/12th scale model of a car that would be going 100km/h, do I test it at 100km/h or does the speed of the wind change? LIke to 1/12th of that or does it need to be faster?

Thanks
-Adam

 

[Airfoil-ed]

Try reynolds number. the formula is on wikipedia. It helps in scaling moving fluids. Funny I came across this site looking for the same thing but found it on wiki

 

[jaap de vries]

You can also put the fan in the back. That is, have a converging section with screens, then have your test section, then have a diffuser (small angle) and then have your fan. This is a pretty standard way for blow (suck) down wind tunnels.

Re = rho * V * L / u or V * L / v

where u and v are the dynamic and kinematic viscosity respectively.

Note that there are two ways to test airfoils, one is having it go across the test section giving you more or less 2 D information. Otherwise make sure the airfoil is not more then 60-80% of the test section diameter to get 3-D measurements

 

[h2oski1326]

You should really look into the topics that FredGarvin mentioned. Dimensional Analysis will play an important factor and simply scaling the Reynolds number may not be enough. Say you are experimentally finding the lift and drag on a airfoil on a small scale model you will likely see other numbers aside from the Reynolds number show up, like an Euler number.

If you are just visualizing the Reynolds number could be sufficient.

The Mythbusters seem to get away with using straw bundles to stabilize the flow from a fan. :)

 

[Airfoil-ed]

I'm attempting both right now; what's the vapor pressure of average air at sea level?
My project is on the effect of winglets on an airfoil. Boeing has them, Airbus doesn't; Why?
(I know the answer is on the internet somewere but I need to apply the scientific method; going for my 5th medal!)

 

[dingpud]

Winglets are the little vertical wings on the end of the main wing right? Aren't they suppose to keep "wash-out air" from coming around the end of the wing thus pushing down on the wing and causing drag? I remember having a conversation about those with someone, but can't remember where it ended up going...

yes, we were in a bar at an airport...

 

[h2oski1326]

(edited)

 

[FredGarvin]

Um. No. You are describing a trip fence. As a matter of fact, winglets can cause separation if they do not have a large enough radius at their point of connection at the tip. They also are moved farther downstream on the wing to a favorable pressure gradient to avoid separation. Do a search on winglets. There are plenty of websites that explain their basic premise.

From NASA's Dryden Research Site:

Winglets increase an aircraft's operating efficiency by reducing what is called induced drag at the tips of the wings. An aircraft's wing is shaped to generate negative pressure on the upper surface and positive pressure on the lower surface as the aircraft moves forward. This unequal pressure creates lift across the upper surface and the aircraft is able to leave the ground and fly.

Unequal pressure, however, also causes air at each wingtip to flow outward along the lower surface, around the tip, and inboard along the upper surface producing a whirlwind of air called a wingtip vortex. The effect of these vortices is increased drag and reduced lift that results in less flight efficiency and higher fuel costs.

ref: http://www.nasa.gov/centers/dryden/about/Organizations/Technology/Facts/TF-2004-15-DFRC.html

Winglets are there to reduce lift induced drag, i.e. induced drag via the disruption of tip vorticies and reduction of spanwise flow. They also help to increase the aspect ratio of the wing. They also use some of the flow from the vorticies to create a small additional forward thrust force. They do increase parasitic drag, add weight, may cause the need for increasing wing stiffness/strength and are pretty darned expensive to retrofit. However, those factors are offset by the fact that you can carry less fuel and have a lower operating costs. It is purely a business case that has to be made to justify their addition to the aircraft. Depending on the aircraft usage, it may not make business sense to have them.

http://en.wikipedia.org/wiki/Wingtip_device
http://www.b737.org.uk/winglets.htm

 

[h2oski1326]

My mistake, you are absolutely correct. I had never heard the term 'winglet' before so I got it confused with a vortex generator.

 

[ank_gl]

in short, winglets give infinite wingspan effect.
But how is a winglet optimized?? there were those 'lil' things back there in past(eg. endplates etc), how is a winglet different from it?? Winglet is, infact, a wing, it also creates its own drag, how is its parameters calculated?? like, what point it should start, what's should be its height.
it makes the craft look cute though