I've lifted the bulk of this out of my "Aero CFD Bible" notebooks I've kept over the years (hence some of the sketches and equation images included below). I think this one was a bit of an amalgamation of a... quorra answer I read once...? Plus some of my own understanding mixed in with it, and whatnot.
Just a small note, I find it a little easier to think in terms of "lift" rather than downforce, so I'll stick to this throughout the post... but the same applies, just in reverse.
▬▬▬▬▬▬▬▬▬▬▬▬▬▬
Aerodynamic Forces Definition
▬▬▬▬▬▬▬▬▬▬▬▬▬▬
So... first of all, lift is only one part of the aerodynamic force in a system. It's the component of aero force vectors acting all over a body, which point "normal" (or perpendicular) to the direction of airflow. Since any aircraft (or F1 car) will distort the local flow around itself, in an ideal situation, you'd want to figure out that direction at an infinite distance away from the thing; where the air is undisturbed. The other (main) component is drag. Similar to lift, it's defined as the component of aero force vectors acting all over a body which point parallel to the direction of airflow.
"Aerodynamic Forces", as a collective term, are defined as the sum of all local pressures (or force vectors over a given area) which act orthogonally (at a right angle) on the local surface of a body; and the shear forces, which act parallel to the local surface.
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
Aerodynamics vs. Electrical Engineering
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
When aerodynamics was first researched into, electrical fields were relatively "new"... and after a few very smart people poked around at the maths and physics behind them, they figured out that those same equations which helped to calcualte electromagnetic field forces could actually be somewhat adapted and then used to explain / calculate aerodynamic forces. This is why in the field of "potential flow theory", you still see terms like "source and sink" used to describe the theoretical model components that make up an aerodynamic calculation. Personally, I don't find electrical engineering interesting at all; day one of electrics 101 at uni saw my uni professor state "electrons go this way... but because one guy made a mistake 500 years ago, we go the other way... and we didn't bother fixing it".
That confusion continues (at least in my opinion heh...) up to the point where these sources and sinks and the mathematics behind them are actually quite complicated and difficult to work with, and not too easy to learn. So as all good engineers (and teachers for that matter) do when something is complicated, we simplify!! Unfortunately, they were mostly too simple and in some cases, just plain wrong... and that kept on being the case for quite a few decades.
Anyways, enough backstory for now, let's move on to the mechanisms at play here:
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
Molecules and Maths
don't freak out, this maths is friendly
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
If we look at things on a molecular level, every air molecule is in a dynamic equilibrium between a few different effects: inertial, pressure, and viscous.
- Inertial, means that the mass of the particle wants to keep moving just as it did before, and you'd need a force to act on it in order to change it
- Pressure, means that the air molecules oscillate all the time and bounce into one another; the more bouncing, the more force they exert on their surroundings
- Viscosity, means that air molecules tend to assume the speed and direction of their neighbouring particles, because of that oscillation stated above
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
Flow over the Upper Surface of the Wing
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
When a wing moves at subsonic speeds, the low pressure area on the upper surface kind of "sucks" in air ahead of it. If we pick some random packet of air for a moment, then as it's travelling towards this wing, then the other air packets above itself, and downstream of itself, there will be "less bouncing around of molecules" because they are being drawn towards this region of low pressure, and so most of their movement will tend in that direction, instead of all around the place; which would imply a lower pressure. Conversely, the air packets below our special chosen one won't have their "bouncing around" diminished, and this has the effect of pushing our packet of air upwards (since it's now higher pressure than the air above our packet) and towards the wing. So our packet (I feel like I'm using "packet" wayyyy too much here now...) will rise and accelerate towards the wing and be sucked up into that low pressure area above the wing.
As the packet accelerates, it gets stretched out lengthways. It's pressure will also drop proportionally to the increase in speed it undergoes, and because it's stretching lengthways, it's also made to contract orthogonally -- it gets thinner. Think of how you'd stretch a lacky band; as you pull it longer, it gets thinner.
As this is all going on, our packet moving over the aerofoil profile suddenly finds that the wing surface below it is curving away from it. If the packet was to keep going in the same direction, then in the "space between it and the wing", there'd almost be like a mini-vaccuum going on. And so, reluctantly (remember, the packet of air has a mass, and therefore an inertia!!) it will change course and follow the contour of the wing profile.
The fast-flowing, low-pressure air will then (in turn) suck on a new air packet ahead of it and below it. This will cause a decelleration and a slight "recovery in pressure" over the second half of the wing's surface. And so the packet has changed direction. Keep that going, and you can get some pretty amazing direction changes... I mentioned in the other post about how adverse pressure gradients, laminar, and turbulent boundary layers work, so I'll just link to that here if you wanna go read up on it: viewtopic.php?p=1055271#p1055271
I think it's pretty intuitive once you've read the above paragraphs, but just in case, lift can only be generated if the upper surface of the wing slopes downwards and away from the initial path of that air packet, once it's navigated around the leading edge. You can accompolish this either by using camber, or increasing the angle of attack of the wing. However, since camber allows for a more gradual change over the surface, it tends to be more efficient than simply increasing angle of attack.
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
Flow over the Lower Surface of the Wing
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
A packet of air which ends up below the wing, will experience less uplift and acceleration than one on the upper side. Also, in the convex part of a highly cambered aerofoil, it will experience a little bit of compression. Even so, it too has to change it's flow direction, because the cambered aerofoils (or angled ones) will push the air below it downwards, creating more pressure and additional "bouncing around" as the packets run into each other. When the packets reach the trailing edge of the aerofoil, they have picked up some downward speed because of this.
Behind the wing, the packets of air continue along their now "changed" (i.e. downward) direction for a short while due to their inertia; and as they do so, they push the air around them downward and sidewards. Above them, the air which has already been pushed sideways before will now move to "fill the void" left by the other packets. Zoom out far enough, this motion looks like two large vortices that have been shed from the surface.
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
Explained in Several Equivalent Ways
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
Looking at the sketch above, one of the ways lift is generated can be expressed as the "delta" (or difference) in pressure between the upper and lower surface of the aerofoil. In terms of molecular "bouncing", the molecules will bounce against the lower side more than the upper.
Or, you could look at a more zoomed out picture: A certain mass has been accelerated downwards by the wing. This required a force to be acted on that mass of air, and it is the reaction of that force in an equal and opposite direction which is lift.
If you were to look at the wing as some kind of "black box" which you only knew what went in, and what came out -- but not what was going on in the middle bit -- the packet of air going in gets some kind of downward "knock" to it, which it reacts with an upward "knock" to the wing, as it exits the black box. In this example, the two paragraphs above explaining what lift is at a molecular level vs. at a macroscopic level would be indistinguishable from one another.
It should be noted that most of this change in direction happens in the first like 30% of the chord of the aerofoil, and not right at the trailing edge.
Additionally, the common-sense assumption about these pressure differentials is that when you want to "make lift", you just increase the pressure below the wing. The reality is that most of your pressure differential between the aerofoil surfaces actually comes from the pressure drop on the suction side of a wing, and not the other way around.
So I guess there is a little bit of deflection... a little induced flow direction change... but of suction... bit of bouncing around... bit of hitting into other things... bit of Newton going on... bit of "off the surface" antics happening... and who knows what else? But one thing is for sure: there's more than one mechanism at play.
▬▬▬▬▬▬▬▬
Supersonic Flow
▬▬▬▬▬▬▬▬
When an aircraft moves faster than pressure changes can propogate through the air around it, the changes in pressure we just talked about are no longer smooht, but sudden. The aircraft will "push" the air molecules aside, producing a compression shockwave. Behind thhe "front" of that shockwave the pressure, temperature, and density are higher than that in front of it; and the delta between them is proportional to the local change in flow direction.
Bear with me for a minute here... this maths is a little scary, but it doesn't bite
The incremental pressure change due to the body hitting air with an incremental angle, expressed in terms of undisturbed (i.e. freestream air) flow, is proportional to the change in streamlines:
Gas pressure on a molecular level is the number, and severity, of particle collisions. The air molecules experience more collisions on the downstream side of the shockwave, since air pressure is higher there. The average direction of those extra collisions is orthogonal to the shock wave because this is the point at which normally-moving air particules suddenly get kicked, jerked, and shoved by the shockwave front. Once a molecule crosses this shockwave, the collisions come from all directions and so any further acceleration is stopped dead in its tracks. If the surface curves away from the local flow direction, the air produces an "expansion fan" which resets the old pressure and density values when the air again returns to it's original direction.
Pure supersonic lift is a matter of "angle of incidence", and basically nothing else. Any local curvature of the wing aerofoil profile will not change overall lift (but it will increase drag). In this case, the total aerodynamic force is normal to the wing, and so drag will also become proportional to the angle of incidence and any extra surface curvature. In hypersonic flow, which is a little out of my usual aerodynamics knowledge trove hah, you basically just use "impact theory", coined by the venerable Isaac Newton.
▬▬▬▬▬▬▬
Separated Flow
▬▬▬▬▬▬▬
Increasing pressure is roughly the same idea as increasing it's potential energy, which leads to a reduction in kinetic energy (from the dynamic pressure portion of the total pressure equation) and therefore an overall average decceleration of the flow.
Since the air right down on the surface, well within the boundary layer, is travelling slower, it is more affected by the adverse pressure gradient serving to deccelerate the flow. And if you continue to deccelerate, meaning over time you're slowing down the average airspeed in that region, you will eventually slow down enough that your velocity hits zero (and then negative). At that point, you're flow is "stalled" or "separated".
Turbulent boundary layers can withstand greater adverse pressure gradients because they have "mixing" within them; rather than laminar layers... so there is energy transport within the boundary "layers". Meaning that the air on the surface, just like with a laminar boundary layer, gets mixed up this time with a bit of higher energy air.
As such, it takes a larger upwind increase in static pressure in order to overcome that mixing effect, slow down the airflow velocity, and then stall the boundary layer. This is why vortex generators are a thing at all; it's because they promote that mixing in the boundary layer by forcing vortical flow regimes at a small enough scale so that they increase the boundary layers ability to withstand the increasing static pressure in the direction it travels (the adverse pressure gradient), which keeps the boundary layer attached for longer: a beneficial effect.
Separated flows remain the only real part of aerodynamics which cannot be precisely predicted, even though its effects are quite well understood. It will still produce lift, but less than that of attached flow (aside from a veeeeeeeery narrow portion of the operating limit where a smidge of trailing edge separation gets you a small net increase in lift if done just right).
▬▬▬▬▬▬▬
Delta Wing Lift
▬▬▬▬▬▬▬
A very specific form of lift generation dominated by flow separation and vortex shape control (with secondary and tertiary vortices too), whilst trying to not cause the vortices to burst. Usually this is referred to as "vortex lift", and it can be quite powerful when used correctly -- particularly at the more supersonic regions of flight.
▬▬▬▬▬▬▬▬
Is this the End...?
▬▬▬▬▬▬▬▬
Yup! That's all folks!
(there's probably a bunch of stuff I've missed but ah well... c'est la vie)
