Lift increasing after stall?

[boyracer]

Hi everyone. Quick question about the coefficient of lift after stalling. Every source I've read on the subject paints a very clear picture about this - the lift force will increase until the angle of attack is increased to an amount such that the air flow separates at the leading edge causing a loss in negative pressure on the upper surface and a loss in lift. As such, once the wing passes the critical angle, it stalls and the coefficient of lift begins to decrease.

However, I've been testing a model of a GOE481 aerofoil that I have made (a very rudimentary model made in the university workshop out of foam with a hot wire), and the coefficient of lift continued to increase right up to 30 degrees when I was expecting it to stall at around 10-15.

There are many things in the data to suggest that the wing has actually stalled at around 15 degrees though. The rate of drag increase takes a sudden increase (ie the curve on the graph becomes suddenly steeper) at this point, and the lift to drag ratio decreases dramatically. Pictures from the smoke tunnel show that the air flow is separating at the leading edge at an angle of attack of 15 degrees as well as turbulent flow towards the trailing edge. The manometers showing the pressure at the leading edge of the underside moved about erratically suggesting turbulent flow. Despite this though, the coefficient of lift continues to increase.

I was wondering if anyone knew of any instances where the coefficient of lift continues to increase after the wing stalls - or perhaps could point me to a reputable source to explain this as I can't find any. My only thoughts are that I've made the span quite short in relation to the chord in order to make it easier to manufacture and fit in the smoke tunnel (chord 360 mm, span 200mm), and that maybe the air is flowing around the sides, but not sure how this would affect anything.

Does anyone have any ideas or suggestions as to how this behaviour could be explained? The McLaren rear wing thread makes interesting reading but I can't find any specific examples in it that could be applied to this situation.

Thank you everyone!

[Mystery Steve]

How are you measuring cL, a 6-DOF balance? Is it calibrated correctly?

Keep in mind that as the aspect ratio decreases, so does the slope of the cL-alpha curve and ultimately the peak cL value decreases as well. However, while the peak value of cL decreases, it will occur at a steeper angle of attack. Raymer talks about this in his textbook, maybe you have a copy in your library. I've sold mine since I took the class, or I would post a figure of what I'm talking about. (I'll rummage through my notes later and see if I can't find something.) Assuming you are calibrated correctly, maybe you're aspect ratio is so small (given the dimensions you stated, it is about 0.5-0.6) that as you're wing is stalling it is at such a high angle of attack that it begins to act as a parachute. Maybe try a larger aspect ratio wing if you're able to manufacture it?

[boyracer]

Thanks for linking to that book. I will have a look to see if its in our library. The span length in manufacture was limited by the size of the hotwire machine so it will be difficult to increase it without manufacturing it by hand which is likely to be quite innacurate. I am going to put some end plates on it however and re-test it to see if it makes any difference.

P.S. I am measuring it using a balance, however I am also using the data taking from the pressure tappings to calculate a coefficient of lift. It produces quite an underestimate due to the way the integral has been approximated to a summation series, but also supports the fact that the Cl keeps increasing.

[Mystery Steve]

... The span length in manufacture was limited by the size of the hotwire machine so it will be difficult to increase it without manufacturing it by hand...

What about decreasing the chord length? Is the machine you're using accurate enough to do this? Maybe you could build the wing as an airfoil approximation and have the span go the width of the test section, maybe by cutting two pieces and joining them?

It all depends on what you're trying to accomplish...

[Raptor22]

you using naked styrofoam?

If yes, have yo considered the effects of the surface finish on the wings performance. are you sure you have boundary layer separation?

[autogyro]

You are forgetting the bottom of the wing, which will continue to create lift up to a steeper angle than the top surface until the bottom flow breaks down and becomes form drag.
It has already been pointed out that there is little to be gained comparing the McLaren wing sections in isolation, if this is what the thread is about.
The air flow directed up from the lower wing section will change the stall angle and stall characteristics radicaly on the upper section and blowing from a slot will increase the angle of stall.

[Tristan]

do you have end-plates on your model?

EDIT: what is the Reynolds number of the experiment?