aerodynamics question

cucumber

hello.

the question is: if i have an exact model (and i mean EXACT) of, say, a glider (you know, those whaddamacallits, white things, long, thin wings, those ones), if i have one of those, why will it not fly as well as a proper one?
(the model is a lot smaller than the real thing)

i was told that the air does not flow around a small wing as well as it does around a large one.
is that true?
if so, why so?

thanks.

turin

I would imagine it has something to do with there being less actual air molecules in a given scaled volume for the model, so the continuum approximation doesn't work as well. Total conjecture, though.

To elaborate on my lines of thinking with an example: Random collisions with individual air molecules will not be as evened out, because there will actually be less (statistically) in any given time. This may amplify random effects.

mathman

Individual molecules are not the problem. The issue has to do with viscosity. The smaller the model, the more the air appears viscous. As a result, the flow is relatively more turbulent, which affects the aerodynamic properties.

Stingray

Look up something called the Reynold's number. Its gives a good rule of thumb for determining when fluid flows are similar.

enigma

With a smaller model you also get reduced lift. Along the tips of the wings, the higher pressure on the bottom of the wing 'bleeds' over to the top of the wing. With a smaller model, there is less surface area to tip area, so the wingtip vorteces have a greater effect.

turin

Why is that?

russ_watters

Enigma and I are very insulted that you posted this in the General Physics forum instead of the Aerospace Engineering forum.

enigma

[:D]

Wha? Er. Oh yeah! VERY insulted...

Integral

You know... You'r right, this is an engineering discussion!

Enjoy!

Phobos

Look up "Scaling Theory" for more info. It's been many years since I studied aero-d, so I forget the details at the moment. [b(] I would need to dust off my old college textbooks. But it sounds like mathman, stingray, and enigma are on the right track. Smaller-scale models are frequently used (wind tunnels, etc.) and there is always some loss of accuracy compared to a full-scale application.

ken

The answer to this lies with the concept of Reynolds number.

Basically due to density and viscosity airflow forms different flows at different speeds and body sizes. Assuming that the density and viscosity are the same, you get the same flow if you double the speed and halve the body size etc. When you scale down a model, you are generally scaling down both the velocity and the size of the body
vastly reducing the reynolds number. The aristream has far less energy to keep laminar flow going long. The curvature of the body is smaller and is much more difficult for the boundary layer to stay attached to the surface. The boundary layer turbulates earlier and is of an much more significant proportion of the body. Any resulting seperation bubbles that would be insignificant on a full size aircraft can extent over a major proportion of the model wing and may in fact not reattach at all.

Reynolds number theory does not hold as good at the vaules related to models because of these scale effects.

skyap

is the scaled model with mass that scaled in the same propotion as the size??

russ_watters

One thing to remember here is that mass is proportional to the cube of any dimension (and area is the square of any dimension). So if you say a plane with half the wingspan of another is half the size - its actually going to have a quarter the wing area and an eighth the volume. Even still, an rc plane is significantly lighter: I googled for a Cessna and a 152 has a wing loading of 10 lb/sq ft and a 172 is 14.5 lb/sq ft whereas an r/c Cessna has a wing loading of about 1.5 lb/sq ft.

One thing to remember is that besides being smaller, the model also flies much slower, generating much less lift per unit area of wing than the real thing. Slower + smaller = that Reynold's number thing discussed above.