What exactly is "stall"
 

What happens to an airfoil at "stall"? And how are "lift" and "drag" defined at high angles of attack?

Imagine a light aircraft. Large wing, low weight, wing strong enough to carry the weight of the plane and safety margin. Assume no thrust - no engines installed or engines shut down.

Absence of thrust means that in steady state, the plane is descending.

The plane might fly at "best glide" AoA. It would then have appreciable forward airspeed, modest rate of descent and good L/D ratio.

If the AoA is any lower, the total airspeed would be higher, L/D would be smaller and rate of descent would be higher.

Now, let´s increase the angle of attack. The forward speed would decrease, and so would L/D. But what happens to RoD?

On the other hand, imagine if the plane is not flying at all: imagine that it has AoA of 90 degrees!

A plane with zero forward airspeed still cannot drop out of the sky at any high speed. After all, as presumed above, it has low weight and large wing. Fast RoD at 90 degrees AoA would mean huge drag. The plane has to reach a steady state at a modest rate of sink and no forward speed. It would be "parachuting" vertically down.

How does the RoD of a parachuting airfoil compare with RoD of a stalling airfoil?

What happens if an airfoil is held at AoA of 80 or 70 or 60 degrees? It should still have a modest RoD - but it should also have a small but nonzero forward speed.

Can someone explain what really changes about the airfoil behaviour if you compare non-vertical parachuting (in "stalled" AoA) with flying "at the back of the power curve", at AoA slightly below "stall"?

My interpretation has been dependent on whether we are discussing airfoil stall, vs. aircraft stall. Bear with me:

An airfoil stall is when the mostly linear (laminar) airflow becomes mostly turbulent. "Mostly" is a subjective term, so instead we look at the classic C-sub-L vs Alpha plot, and somewhere around or slightly beyond the peak lift value, we call it stalled.

The aircraft stall is often called the "break", and it is when the wing no longer develops enough lift to balance the tail surface, thus a nose-down pitch couple develops.

The two may not necessarily be exactly the same.

We used to call this a "falling leaf" manuver. In a light plane you could hold it just above the stall break, keep the ailerons centered, and keep wings level using the rudder. It was unstable in this attitude, so you were constantly re-leveling it, and if you didn't, it might fall off in a spin. But it could be done. Here is a link to a forum.

If one were desperate (or foolhardy) enough I guess you could maintain this down to a couple hundred feet, then drop the nose a bit for a short-field landing. :eek:

Well, of course stability is another issue... Some planes seem to have the problem that "falling leaf" stall is unstable and tends to turn into a spin - others have the problem that stall is a stable "deep stall"... A common trouble seems to be that spin is also stable.

But it seems that simply because of air resistance, neither stall nor spin can be the fastest way down. The fastest (and thus the most dangerous) should be dive.

I occasionally demonstrate this maneouver to students with strong tendencies to use aileron to correct for a wing-drop during practice stalls, to simply demonstrate the effectiveness of the rudder.

In you typical cessna (152 and 172) this will only yield a descent rate of ~500fpm or so. No screaming-ass-descent at least.

Then there's the question how stalled this condition really in a forgiving trainer with a forward CG. Kinda goes in and out of being stalled even with full elevator back pressure.

Not sure how a PA-28 would behave - is the rudder in clean airflow? Cessnas have a good bit of rudder below the horizontal tail surfaces - but many types don't.

:confused:

I find it easier to build a mental picture of this stuff if I focus on the wing, ignoring the aircraft. An aerofoil in a wind tunnel for example.

Lift is a force acting at right angles to the relative airflow. Drag is a force opposing motion. Together they make up the total effect of the airflow on the wing. This definition is just an engineer's way of breaking the very complicated total aerodynamic force on the wing into two easily defined components.

As the angle of attack increases (speed etc unchanging), lift increases steadily and drag increases gently. The airflow will be mostly following the upper surface of the wing as barit1 said. At some point the lift will stop increasing, drag will be increasing more rapidly, and the airflow will (quite quickly but not instantly) go from mostly following the wing's surface, to mostly separated from the wing.

If you hold the back of a spoon under running water the water will tend to follow the curve of the spoon. But if you tilt the spoon more and more so the back side of the spoon is underneath, at some point the water will say "bugger this, I'm not going to stick to the spoon any more, I'm just going to go straight past it". You have exceeded the spoon's critical angle of attack in water, and lift and drag will change quite quickly.

In the former you have a large aerodynamic force that is mostly drag and very little (but still some) lift.
In the latter you have a large aerodynamic force that is mostly lift with a reasonably sized drag component as well.
The difference is just whether or not the airflow is mainly sticking to the surface of the wing (allowing the wing to do its job efficiently) or whether the airflow is mainly just going straight past the wing without sticking to the surface, in which case you'll get lots of turbulence and not much efficiency.

Whether the aircraft is descending and its ROD depends on other factors, like weight, centre of gravity, thrust, how effective the horizontal stabiliser is, what the pilot is doing with the elevator, etc.

Yes, that's true. But in a dive the pilot has full control over the flightpath and limited control over speed. In a stalled condition the pilot has limited control over flightpath and no control at all over speed, as the aerodynamic surfaces will be inefficient until the aircraft is "unstalled". Surely it is more important to consider how easily you can stop descending rather than the actual ROD?

You can check http://www.youtube.com/watch?v=6-M9W7vGfNw&NR to see a Su-35 flying at greater than 90 AOA.

Stall is to do with momentum transfer within the boundary layer, in turn a function of fluid viscosity.

As the fluid within the boundary layer closest to the surface decelerates, it transitions from laminar to turbulent and then reversed flow.

The point of separation is normally associated with this point of flow reversal.

When separation first occurs over a significant part of the wing, it is said to be stalled.

Some aspects of turbulent flow are contra-intuitive.

For instance, turbulence delays separation because the mixing essentially re-energizes the boundary layer fluid, enabling it to combat the adverse pressure gradient - hence the presence of vortex generators on some wings.

However, turbulence obviously increases skin friction, hence the research into laminar flow wings.

With or without thrust?
Is a forward CG all that forgiving? Inverted stall of tailplane should lead the craft into a nosedive, which is the fastest way down...

SR71
Thank you for that. Could not have put it better myself!
As a total nit-pick and considering the context of this thread which is on wing aerodynamics and the stall you could have put ‘operating under full control’ in place of ‘flying’ after your youtube link.
JF

Early C-177's had this characteristic, hence the LE slot (inverted) on the horizontal stabilizer.

However, if there is no thrust, the only way is down. The plane can rise briefly by exchanging airspeed for altitude, but a steady flight at constant speed id necessarily descending. It is the rate of descent which can vary.
This, of course, depends on exactly how the aerodynamic controls work. The one condition where they definitely are inefficient is at standstill, and as there are no aerodynamic forces at all at standstill, is is not a sustained condition.

But yes, many low subsonic planes have the problem that stall is stable while dive is not. In high subsonic, of course, dive tends to be stable as well due to Mach tuck...

Hi again chorned,

You've had three different explanations of the stall from barit1, me, and SR71 - I'm not sure what it is you don't understand.

If you are seeking to prove that a stalled descent occurs at a lesser ROD than a dive, you will not find many people to argue with you. You may be right (depending on the design of the particular aircraft), but only pilots doing aerobatics will deliberately descend this way, because the aircraft is not fully controllable when stalled.

Except for the SU35 naturally... :ok:

On a different subject, while the SU35 is fully controllable in pitch, roll and presumably yaw during its high AoA manoeuvre, I have seen a real-life example, and it appears the pilot has very limited control over the velocity vector during the manoeuvre. Is this the case SR71 or JF?

[quote=SR71;3051708]You can check http://www.youtube.com/watch?v=6-M9W7vGfNw&NR to see a Su-35 flying at greater than 90 AOA.

This seems to me to be a good example of what the great Alan Branson (spelling?) called "thrust-supported contraptions".

OK let's do the algebra, for a glider (ignore thrust).

Basic equations are

L = W * cos(q)
D = W * sin(q)

where q is the angle of descent. If we further look at L and D in terms of the coefficients:

L = 1/2 * rho * v^2 * A * Cl = W * cos(q) (1)

D = 1/2 * rho * v^2 * A * Cd = W * sin(q) (2)

So in the general case, there's a pair of simultaneous equations in v and q. We have parameters like rho, A and W (which I won't mention again). If we assume that the controls are powerful enough for us to choose an AoA and therefore the corresponding Cl and Cd, then Cl and Cd are just more parameters.

What we're accustomed to saying for flight is that q is small so we assume cos(q) is 1, and we use equation 1 to determine v vs Cl on the assumption thatit's lifting its own weight W. But the question chornedsnorkack is asking is, "what happens when we go past stall AoA and Cl is no longer high enough to allow it to support its weight W?" It's counterintuitive to expect it to speed up to allow more v to make up for less Cl, particularly as the AOA gets higher and higher and Cl therefore gets very low.

In fact, it doesn't have to. All it has to support is W * cos(q) and by the time we're in a deep stall, q is high enough so that cos(q) is a lot less than 1. The remainder of the weight is now opposed by drag. And in the extreme case, where the aircraft is going straight down, has q = 90 degrees and all the weight being opposed by drag.

So the solution to the equations is that

tan(q) = Cd/Cl (which is what you'd always expect)

but then solving for the speed you get

1/2 * rho * v^2 * A * sqrt(Cl^2 + Cd^2) = W

so the drag is helping the lift out in opposing the weight.

I'd love to give a real numerical example but unfortunately most of the usual sources don't quote drag beyond the stall. So let's pick some numbers. Imagine that at stall we manage Cl = 1.5 and Cd = 0.3. So our glide angle is arctan(0.3/1.5) about 11 degrees.

Now we go beyond stall. Cl falls to 1. Cd rises to 1. Now our glide angle is 45 degrees. If you plug the numbers in, you get a speed about 4% above stall speed, so we're plummeting a little quicker than when we were in lifting flight, but nothing dramatic has changed except the glide angle.

Now let's put the wing at an AoA of 90 degrees. Cl falls to zero, Cd rises to 2. Our glide angle is 90 degrees (straight down). Our speed falls to 87% of stall speed.

Does that help?

So, anyone actually stalled with no premeditation?

It is important to remember that a stall can occur at ANY airspeed [including a high speed dive-with full power]-the aircraft would NOT survive.
a stall results from excessive AoA it can occur straight up or straight down or otherwise as long as your AoA control (stick or wheel is too far back) and you present an excessive angle of attack to the Relative Wind You can stall at any pitch attitude.
In my honest opinion---All airmen should acquaint themselves with the V-g or V-n diagram and really appreciate the adverse effects of stalls above the maneuver speed

please give due regard to this fact, it is extremely important and can kill you

Well, just to fan the flames...

When landing a hang-glider in nil wind, the pilot initiates a snap stall to halt the forward speed of his contraption and gently:rolleyes: :{ alight on his feet.
The snap stall is achieved by rotating the entire wing from a relatively normal flying attitude (AoA) to the vertical. IE AoA about 90 degrees. This all takes place within a fraction of a second, all without thrust (except gravity of course) and when carried out at the right airspeed is a perfect solution to landing without wheels from an approach speed of about 20+mph.
Can't see it working on an a/c with a high wing loading mind.:ugh:

Snap stall - steeply pitching up to arrest forward movement and gently sink on feet - is also SOP for birds. Who have a modest wing loading as well.

Note that many planes have poor L/D ratios even at best glide. While high-performance gliders have best glide L/D over 60, and subsonic airliners in the range of 15...20, Concorde has best glide L/D under 11, and on landings and takeoffs at the back of power curve is said to have as little as around 4. L/D in the range of 3...4 is said to be common for ultralights - at best glide. They would do worse than that in the conditions of back of power curve.

If a plane is, say, moving forwards at 24 knots and at the same time descending at 10 knots, what is the relevant airspeed? 24 knots, or 26 knots?

Lift, drag and AoA are all referenced to the relative airflow. So is airspeed. The wing doesn't care if a component of the airflow is in a direction that you happen to have labelled as "down" or "forwards".

Sort of.....

Auster Autocrat, decades ago, young, inexperienced and stupid, low flying 100' -200' in a club handicap race (you could do that in those days), at a turn mark behind someone, rolled into 160 turn as he did, full power and tightened it hard to get round closer to the mark than he was, and thus get in front of him. In an Autocrat you needed to do something like that to win.

Felt the stall begin without really realising what was happening, reduced turn as kneejerk reaction, thereby very nearly hit the other guy as he turned, abandoned race, shaking all over. Went home very quietly.

The point has already been made..speed had nothing to do with it; it was more that twice, even three times the stalling speed of that aircraft in level flight, light load, no flap, power off.

I've never been able to work out what the sequence would have been if it had stalled completely, but it would have been terminal and quick.