The first time a wing foil board lifts out of the water, the noise is what you notice. One second the hull is slapping through chop and the spray is in your face. The next second it stops. The board is floating half a metre above the surface, and the only thing still touching the sea is a carbon mast about as thick as a ruler.

Under that mast there is a wing. It is roughly the size of a sheet of A3 paper, and it holds up a person, a board and all the gear while it moves at the pace of a fast bicycle.

In this article I want to explain how that wing works. We’ll look at what the water does as it flows around the foil, why that flow creates lift, how a rider controls it with nothing but their feet, and what eventually limits how high and how fast a foil can fly. Most of the ideas are the same ones that keep aeroplanes in the air. The water changes the numbers, and it adds a few problems that aeroplanes never have.

Here’s the whole system. Drag the slider to change the board’s speed and watch the forces on the rider:

There’s a speed at which the foil’s lift reaches the rider’s weight, and the board leaves the water. Below that speed, buoyancy makes up the difference. Above it, the foil could lift more than needed, so the rider has to reduce lift to stop climbing. That last part may sound odd, but it is the core of riding a foil, and we’ll come back to it.

Water is heavy #

Before looking at any flow, it’s worth asking why the foil can be so small. An aeroplane carrying the same 92 kilograms would need a wing the size of a room.

The answer is density. A cubic metre of seawater has a mass of about 1025 kg. The same cubic metre of air has about 1.2 kg. Water is roughly 840 times denser, and the force a wing produces scales directly with the density of the fluid it moves through. Every litre of water the foil pushes around carries 840 times more momentum than the same litre of air would.

The demonstration below compares the wing area you’d need to lift the same rider at the same speed, in air and in water, drawn to the same scale:

At wing foiling speeds, the air wing would span tens of metres. The water wing spans less than a metre. That’s the whole trick of the sport in one picture: water is dense enough that a small, light, portable wing can do the work of a glider.

Watching water move #

Water is mostly transparent, and even when it isn’t, you can’t see how it moves around a foil half a metre down. To reason about it we need a way to make the flow visible.

The first decision is where to stand. From the beach, the foil moves through still water. From the board, the foil is still and the water streams past it. Physically, the two pictures are identical: the forces only depend on how the water and the foil move relative to each other. Switch between the two views:

Seen from the beach, the water only stirs as the foil passes and then settles again. Riding along, it’s much easier to see what the foil does to the flow, so for the rest of this article we’ll sit on the foil and let the water come to us.

One more simplification: I’ll mostly show a cross-section of the wing, a slice cut along the direction of travel. This shape is the foil’s profile, and in two dimensions it captures most of what matters. The long, swept, three-dimensional shape of a real wing will come back later when we talk about drag.

There are a few ways to draw flow. Arrows at fixed points show the water’s direction and speed at each spot. Markers are tiny parcels of water that drift with the flow and leave a short trail. A speed map colours every point by how fast the water moves there. Switch between them:

The flow here is steady: the arrows never change, even though the markers keep moving. Every marker that starts in the same place follows the same path.

A few things stand out. The water slows to almost nothing right at the nose, splits, and speeds up as it wraps over the top. The water passing underneath is slower. Release a line of markers in front of the foil and watch closely: the markers that go over the top reach the back of the foil well before their neighbours that went underneath. They don’t meet up again. The popular explanation of lift says the two halves must arrive at the trailing edge together, so the top flow has to go faster. That’s simply not what happens. The top flow is faster, but not for that reason.

A note on these simulations. They use a classic mathematical model of an ideal fluid, one with no friction at all, flowing in two dimensions. It’s good at showing the pattern of the flow around a foil, and it gets the lift of a thin profile at small angles surprisingly right. It’s wrong about drag and about what happens near the surface of the foil, which we’ll fix with words where it matters. Please don’t design a foil with it.

Pressure #

Water pushes on everything it touches. That push per unit of area is pressure, and it comes from countless molecules colliding with a surface. Half a metre underwater the pressure is about 106 kilopascals: the atmosphere above plus the weight of the water column.

That’s a large number, but on its own it does nothing to the foil. The water presses on the top and on the bottom, and if the pressure were the same everywhere, the pushes would cancel. A foil only feels a net force if the pressure on one side is different from the other.

So rather than show absolute pressure, I’ll show the difference from the ambient pressure at that depth. Red means the water is pushing harder than ambient. Blue means it’s pushing less:

Compare this map with the speed view from before. Where the water is fast, the pressure is low. Where it slows, at the nose, the pressure is high. This is Bernoulli’s principle, and the cause and effect is easy to get backwards. Water is pushed from high pressure towards low pressure. A parcel of water moving into a low-pressure region gets pushed from behind harder than from in front, so it speeds up. The pressure field and the speed field are two descriptions of one flow, and neither one comes first.

In the forces view, the blue arrows on the top surface point outward: the low pressure is effectively pulling the skin of the foil upward. The red arrows on the bottom push inward and up. At normal angles, the top surface does most of the work. Most of the lift comes from suction over the top, not from water bouncing off the underside.

There’s a second way to describe the same force. The foil turns the flow slightly downward as it passes. Newton’s third law says that if the foil pushes water down, the water pushes the foil up. The pressure picture and the momentum picture are the same lift described in two ways.

Angle of attack #

Try the experiment everyone has done: put your hand out of a moving car window, palm flat. Tilt the front of your hand up a little and you feel it lift. Tilt it more and it lifts more. Tilt it too far and the lift suddenly turns into a messy, shaking drag.

The tilt is the angle of attack: the angle between the foil’s chord line, a straight line from nose to tail, and the oncoming water. It’s the single most important variable in foiling. Lift grows almost in proportion to it, until it doesn’t:

For small angles, lift rises in a straight line. A cambered wing, one with a curved profile like this one, produces lift even at zero degrees, which is why the line crosses zero at a slightly negative angle. Normal riding happens between about 2° and 8°.

Past roughly 12–15°, the water flowing over the top can no longer follow the curve of the upper surface. To stay attached, it would have to keep flowing into rising pressure towards the trailing edge, and at some point it runs out of momentum and peels away. This is stall. Lift drops, drag shoots up, and on a board the result is immediate: the foil stops carrying you and you drop.

On the water, you change the angle of attack with your feet. Weight on the back foot lifts the nose of the board, which tilts the whole foil nose up and increases the angle. Weight on the front foot does the opposite. The car-window hand is the front wing, and your feet are the wrist.

The lift equation #

All of this condenses into one formula that every foiler meets eventually:

L = ½ · ρ · v² · C_L · A

L is lift. ρ (rho) is the density of the water, the 1025 kg/m³ from earlier. v is your speed through the water. C_L, the lift coefficient, is the number from the previous chart: it bundles up the foil’s shape and angle of attack into a single value, usually between 0.3 and 1.2 when riding. A is the area of the front wing.

The term to stare at is v². Lift grows with the square of speed. Double the speed and lift quadruples. This is why takeoff feels abrupt: near the threshold, a small gain in speed produces a large gain in lift, and the board pops up rather than rising gently.

Play with the inputs and watch where the lift curve crosses the rider’s weight:

With a 92 kg total load, a 1200 cm² front wing and a lift coefficient of 0.8, takeoff comes at about 15 km/h. Rearrange the formula and two practical rules fall out. A bigger wing takes off at a lower speed, which is why beginners learn on wings of 1500–2500 cm². A heavier rider needs more lift, so they need a bigger wing or more speed. Racing wings of 700 cm² or less need 18 km/h or more before they fly at all.

Once you’re flying, the same formula explains what the rider does with their feet. Speed keeps rising, so lift would keep rising too, and the rider has to keep reducing the angle of attack to hold it equal to their weight. In the first demo, look at the angle the rider holds at 25 km/h: barely one degree.

Flying the board #

A foil is not just a front wing. Behind it, at the end of a horizontal bar called the fuselage, sits a much smaller wing called the stabilizer. It works like the tail of an aeroplane.

The stabilizer is set at a slightly lower angle than the front wing and pushes gently down. If something tips the nose up, the front wing’s angle increases, but so does the stabilizer’s downward force, and the stabilizer acts at the end of a long lever, the fuselage. The extra push at the back levers the nose down again. The system corrects itself.

The longer the lever, the stronger that correction. That’s the trade-off behind fuselage length: long fuselages are calm and forgiving, short ones respond quickly and feel agile.

Try to fly the board below. Shift the rider’s weight to control the pitch, hit a bump to disturb it, and compare a short fuselage with a long one:

A few things to notice. Small weight shifts produce surprisingly large changes in height, because every degree of angle changes lift by more than ten percent. With a short fuselage, the same weight shift pitches the board further, and a bump makes it bob for longer. And if you keep climbing, the foil eventually gets close enough to the surface to lose lift. We’ll see why in a moment.

Roll works on a similar principle. Many front wings have their tips angled slightly upward, called dihedral. When the foil rolls, the deeper tip sits in higher-pressure water and lifts more, and the shallower tip lifts less, which rolls the foil back to level. Wings with flat or drooping tips give up some of that self-correction in exchange for quicker carving.

Drag, and why foiling feels like cheating #

Lift is never free. Anything moving through water also feels drag, a force pulling it backward. When the board is in the water, most of that drag comes from two places. The hull makes waves, and those waves carry energy away. The hull also rubs against the water over a large wetted area.

When the board lifts off, both of those disappear. Only the mast and the foil stay in the water, and they’re slim and streamlined. This is why foiling feels effortless once you’re up: drag drops by somewhere around 60–80%.

What remains is the drag of the foil itself. Part of it is profile drag: friction from water sliding along the surface, plus the pressure cost of the wake behind the foil. The rest is induced drag, and it comes from the fact that a real wing has ends.

High pressure under the wing and low pressure above it want to equalise, and at the tips they can. Water curls around each tip from bottom to top and trails behind as a pair of spinning tip vortices. Making those vortices costs energy, and that energy is drag. The cost depends heavily on the wing’s aspect ratio, its span divided by its average chord. A long, narrow wing spreads its lift across a wider stretch of water and loses less of it at the tips:

At an aspect ratio of 4 or 5, the shape of a beginner wing, induced drag is the larger part. At 10 or more, the shape of a racing wing, it shrinks a lot and the lift-to-drag ratio climbs. Good high-aspect wings reach 25 or more: 25 units of lift for every unit of drag, comparable to a glider.

The cost shows up elsewhere. Long, thin wings are harder to build stiff enough, react faster to every input, and forgive less. Short, wide wings are stable and stall gently. There’s no best aspect ratio, only a best one for a rider and their conditions.

Shaping the foil #

Back to the profile. Two numbers describe most of its personality.

Thickness is the deepest point of the profile as a fraction of its chord. Thick profiles, around 14–18%, are strong, tolerate higher angles before stalling, and are forgiving. Thin profiles, around 8–10%, have less drag but need carbon fibre to stay stiff and give the rider less margin.

Camber is how curved the profile’s centreline is. A symmetric profile, with no camber, produces no lift at zero angle of attack, because the top and bottom are mirror images. A cambered profile bends the flow even when it’s level, so it lifts at zero degrees and lifts more at every angle.

Play with both and watch what happens to the pressure field and to the lift at 0°:

A stabilizer is often close to symmetric, because its job is to trim and steady the foil rather than to lift. A front wing is cambered. A wing built for light wind is heavily cambered, so it lifts early and lets you take off at low speed, and it pays for that with more drag once you’re fast.

Recent research from MARIN, the Dutch maritime research institute, put numbers on how much each geometric choice matters. Scholcz and colleagues trained a machine-learning model on a large set of flow simulations of parameterised hydrofoils. They found that camber has the largest effect on both lift and drag, larger than thickness, taper or the shape of the planform. If you’re choosing a foil, the curve of its profile is the first thing that shapes how it rides.

A rough guide to how all these choices combine:

The ceiling above: the surface #

An aeroplane wing flies in a sky with no lid. A hydrofoil flies under one. The surface of the water sits just above it, and that changes things.

Above a foil there’s a region of low pressure. Deep down, the surrounding water holds that low-pressure region in place. Near the surface, the water above the foil is free to move, so the surface dips toward the low pressure and leaves a train of waves behind. Making those waves costs energy, and the distortion steals some of the lift.

Two numbers describe how bad this gets. The submergence ratio is the foil’s depth divided by its chord. Below about one chord, the surface dominates. Beyond about three chords, the foil behaves almost as if it were in open water. The Froude number compares the foil’s speed with the speed of the waves it makes. At low Froude numbers, the foil stays under the big waves it creates and loses a lot of energy to them. At high Froude numbers it outruns them.

Bring the foil close enough and something worse happens. The low pressure on the top surface sucks air down from the surface, and the air spreads over the foil. This is ventilation. A wing covered in air has nothing to push against, so lift collapses almost instantly. For intermediate riders, ventilation is the most common cause of crashes: riding too high, or carving so hard that one wingtip comes near the surface.

This also explains why takeoff is the hardest moment. At takeoff you’re slow, so the Froude number is low. The foil is still close to the surface, so the submergence ratio is low. And at low speeds the junction between the mast and the wing disturbs the flow more than you’d expect: one study cited by the MARIN team measured a T-shaped foil producing only about 61% of the lift slope theory predicts. Every penalty arrives at once, exactly when you need the most lift. It’s one more reason learning foils are big.

The ceiling ahead: cavitation #

The final limit is the strangest. Water boils when its pressure drops below its vapour pressure, about 2.3 kilopascals at 20 °C. This is the same thing that happens at the top of a mountain, where water boils below 100 °C. Push the pressure low enough and water boils at sea temperature.

The pressure on a foil’s suction side drops further as speed rises, and it falls with the square of speed, like lift. Go fast enough and the lowest-pressure spot on the foil crosses the vapour pressure. Bubbles of vapour form, are carried along, and collapse violently as they reach higher pressure again. This is cavitation. Riders hear it as a crackle or hiss, feel it as vibration, and over time it pits the surface of the foil.

Whether a foil cavitates depends on two things: how fast it’s going and how hard it’s working. A foil at a high angle of attack has a sharp suction peak at its nose and cavitates there first. At a strongly negative angle, the peak moves to the underside of the nose. Between those extremes there’s a window of safe angles, and that window narrows as speed rises.

Plotting the window for every angle and every speed produces a shape engineers call the cavitation bucket. Below, speed increases downward. Everything above the black line is clean flow:

At ordinary speeds, the bucket is wide and you can ride sloppily. As speed climbs, the walls close in. At racing speeds, a degree or two of extra angle is enough to start cavitating. Thicker profiles widen the bucket and forgive more. Camber shifts it toward positive angles, which suits a front wing that always works at positive angles. Most recreational riders never reach cavitation, because it starts somewhere in the 25–40 knot range for typical foils. It’s the reason speed-record foils look so different from everyday ones.

The model behind this chart treats the nose of the foil more harshly than real flow would, so the exact numbers are rough. The shape is the real thing.

Water is slightly sticky #

I’ve been ignoring one property of water this whole time: viscosity, its internal friction. Right at the foil’s skin, water sticks to the surface and doesn’t move at all. A few millimetres away it moves at full speed. That thin region of sheared water is the boundary layer, and it’s where friction drag and stall really come from.

How a boundary layer behaves depends on the Reynolds number, a ratio of how strongly the flow’s momentum carries it along versus how strongly viscosity smooths it out. Hydrofoils live at Reynolds numbers between about 100,000 and 1,000,000. That’s an awkward middle ground where the boundary layer can be smooth and orderly (laminar) or chaotic (turbulent), and it can switch between the two unpredictably.

Laminar flow has less friction but separates more easily. Turbulent flow costs a little more friction but holds on to the surface better. Some performance foils add a carefully placed strip of roughness to force the switch at a known spot, trading a little drag for predictable behaviour. It also means a deep scratch on your front wing is more than cosmetic. It changes where the flow switches and how the foil behaves near stall.

References #

Research papers #

1. Scholcz, T.P., Odendaal, K., Marelli, G., Minerva, L., Schnitzeler, O.V., van Walree, F. (2026). Development of an early stage hydrofoil design tool including unsteady hydrodynamics. Applied Ocean Research 169, 105011. doi:10.1016/j.apor.2026.105011. Used for camber as the most influential geometry parameter for lift and drag, the non-dimensional geometry ranges, the submergence ratio and Froude number regimes, and the cavitation bucket regions and their risk ranking.
2. AlaviMehr, J., Davis, M.R., Lavroff, J. (2015). Low Reynolds number performance of a model scale T-foil. International Journal of Maritime Engineering 157 (A3). doi:10.3940/rina.ijme.2015.a3.336. Used for the T-foil reaching only 61% of the ideal lift curve slope at low Reynolds numbers, because of the vertical strut.
3. Bosschers, J., Vaz, G., van Wijngaarden, E. (2008). Computational analysis of propeller sheet cavitation and propeller-ship interaction. Semantic Scholar. The method Scholcz et al. use to compute cavitation buckets, including sheet cavitation at the leading edge at high angles and bubble cavitation mid-chord at small angles.

Free surface studies (cited by Scholcz et al.) #

These are the studies Scholcz et al. cite for the claim that a submerged foil’s performance depends on its Froude number, its depth relative to its chord and its angle of attack. They support the section on the surface above the foil.

4. Pernod, L., Sacher, M., Wackers, J., Augier, B., Bot, P. (2023). Free-surface effects on two-dimensional hydrofoils by RANS-VOF simulations. Journal of Sailing Technology 8 (01), 24–38. doi:10.5957/jst/2023.8.2.24.
5. Nicolas, H., Perali, P., Sacher, M., Bot, P. (2023). Boundary element method analysis of 3D effects and free-surface proximity on hydrofoil lift and drag coefficients in varied operating conditions. Journal of Sailing Technology 8 (01), 183–199. doi:10.5957/jst/2023.8.10.183.
6. Perali, P., Sacher, M., Leroux, J.-B., Wackers, J., Augier, B., Hauville, F., Bot, P. (2024). Performance prediction of a hydrofoil near the free surface using low (BEM) and high (RANS) fidelity methods. Applied Ocean Research 151, 104157. doi:10.1016/j.apor.2024.104157.
7. Bot, P., Hauville, F., Ploé, P., Longuet, J. (2025). Tank testing of a windfoil hydrofoil with free surface effects. Journal of Sailing Technology 10 (01), 191–205. doi:10.5957/jst/2025.10.1.191.
8. Esmaeilifar, E., Hassan Djavareshkian, M., Forouzi Feshalami, B., Esmaeili, A. (2017). Hydrodynamic simulation of an oscillating hydrofoil near free surface in critical unsteady parameter. Ocean Engineering 141, 227–236. doi:10.1016/j.oceaneng.2017.06.037.

Background #

The remaining material is standard aerodynamics applied to an underwater wing: the lift equation, the pressure and momentum views of lift, angle of attack and stall, the takeoff speed calculation, pitch and roll stability, the kinds of drag, aspect ratio and induced drag, ventilation and the Reynolds number regime. It comes from my research notes on these topics, which summarise textbook theory rather than any single paper. Typical figures such as wing areas, takeoff speeds, stall angles, the 60–80% drag reduction and the drag budget chart are representative values, not measurements.

Final words #

When I’m up on a foil, I don’t think about pressure fields or tip vortices. I think about the next swell and where my back foot is. But all of it is happening under me every second: water speeding up over the top of a small curved wing, a pressure difference of a few kilopascals spread over an A3-sized surface, a stabilizer levering the nose down, and the surface above and the vapour pressure ahead marking the edges of what’s possible.

The quiet you notice when the board lifts off is what all of that sounds like.