On the ventilation of surface-piercing hydrofoils under steady-state conditions

This experimental study investigates the onset of ventilation in surface-piercing hydrofoils under steady-state conditions, identifying three distinct trigger mechanisms and proposing a revised stability map that reveals non-uniform boundaries between flow regimes and higher critical angles of attack than previously estimated.

The Gist

Imagine a hydrofoil as a high-speed underwater wing attached to a boat. As the boat speeds up, this wing lifts the hull out of the water, reducing drag and making the ride faster and smoother. However, there's a tricky problem: if the wing goes too deep or hits the water at the wrong angle, air can get sucked down from the surface, creating a giant bubble around the wing. This is called ventilation. When this happens, the wing loses its grip on the water, the lift disappears, and the boat can suddenly drop or shake violently.

This paper is like a detective story trying to figure out exactly when and how this air bubble forms on these underwater wings.

The Two Ways to Test the Wing

Scientists usually try to predict when ventilation happens by looking at two main things: how fast the boat is going (Froude number) and the angle the wing is tilted (angle of attack).

In the past, researchers mostly did experiments like this:

1. The "Speed Up" Method: They would set the wing at a fixed tilt and then slowly speed up the boat until the air bubble appeared.
2. The "Tilt Up" Method (This Study): The authors tried a different approach. They set the boat to a specific speed and then slowly tilted the wing up until the air bubble appeared.

They found that these two methods give very different answers. It's like trying to find the edge of a cliff. If you walk straight toward it (speeding up), you might fall off at one spot. If you walk sideways along the cliff edge (tilting up), you might find the edge is actually much further out than you thought.

The Three Ways Air Sneaks In

The researchers discovered that air doesn't just "suck in" the same way every time. Depending on the speed and the shape of the wing, air uses three different "backdoors" to get inside:

1. Nose Ventilation (The Front Door):

   • When it happens: At lower speeds.
   • How it works: Imagine water flowing over the front of the wing. At certain angles, the water slows down and swirls in a little pocket (a bubble) right near the front edge. This pocket creates a vacuum. If the water layer covering this pocket gets too thin, the air from the surface pokes through like a needle popping a balloon.
   • The Result: This happens quickly (in about 3.5 seconds of "wing time"). It's the most common way air gets in at lower speeds.

2. Tail Ventilation (The Back Door):

   • When it happens: At higher speeds.
   • How it works: As the wing moves fast, it pushes the water down. This creates a kind of "downward wind" effect on the water surface behind the wing. Tiny ripples on the water get stretched and pulled down so hard that they turn into air-filled tornadoes. These tornadoes grow until they connect the surface air to the low-pressure area under the wing.
   • The Result: This is a slower, more gradual process (about 7 seconds of "wing time"). It takes over as the main way air gets in when the boat is going fast.

3. Base Ventilation (The Side Door):

   • When it happens: Only on wings with a flat, blunt back end (like a semi-ogive shape).
   • How it works: Air tries to sneak in through the wake (the trail of water) right behind the wing.
   • The Result: The researchers found this didn't actually create a stable, dangerous bubble in their tests. It was more like a false alarm or a precursor to the "Tail Ventilation" method.

The Big Surprise: The "Safety Zone" is Bigger Than We Thought

The most important finding is about the Stability Map. Think of this map as a weather forecast for the wing, telling you when it's safe (fully wet) and when it's dangerous (ventilated).

• Old Map: Previous studies suggested that if you tilt the wing past 15 degrees, it would almost immediately lose its grip and get ventilated.
• New Map: The authors found that if you approach the problem by slowly tilting the wing (rather than speeding up), the wing can actually handle tilts up to 25 degrees or more without getting ventilated!

This means the "danger zone" is much smaller than we thought, but only if you approach it carefully. The old maps were missing a huge "safe zone" because the way they were testing (speeding up) forced the air to enter earlier than it naturally would if you just tilted the wing.

Why Does This Matter?

The paper explains that the shape of the wing matters. Thin wings are prone to the "Nose Ventilation" (front door) trick, which happens at lower speeds. Thicker, sturdier wings might avoid this trick entirely, allowing them to stay stable at even higher speeds and angles.

In summary: The researchers showed that the rules for when a hydrofoil loses its grip on the water depend heavily on how you get there. By slowly tilting the wing instead of just speeding up, they found that the wing is much more stable and can handle steeper angles than previously believed. They also identified that air uses different "tricks" (front, back, or side) to get inside depending on how fast the boat is going.

Technical Summary

Problem Statement
Surface-piercing hydrofoils are critical for high-performance marine vessels due to their ability to reduce drag by lifting the hull out of the water. However, their interaction with the free surface makes them susceptible to ventilation, a phenomenon where air is drawn from the atmosphere into the low-pressure region on the suction side of the foil. This leads to a sudden loss of lift and potential structural instability. While the conditions for ventilation onset are generally understood to depend on the angle of attack ($\alpha$) and the depth-based Froude number ($Fr$), accurately mapping the boundaries between flow regimes (fully wetted, partially ventilated, and fully ventilated) remains challenging. Previous studies, notably Harwood et al. (2016), have produced stability maps based on a "conventional" approach where the Froude number is varied at a fixed angle of attack. However, the assumption that the resulting stability map is independent of the surveying procedure (i.e., the path taken through the parameter space) has remained largely untested. Furthermore, the specific trigger mechanisms initiating ventilation and their dependence on hydrofoil geometry and flow conditions require further elucidation.

Methodology
The study was conducted experimentally in the towing tank of the Ship Hydromechanics Laboratory at Delft University of Technology. Two simplified model hydrofoils were tested: a semi-ogive profile with a blunt trailing edge (matching the geometry used by Harwood et al., 2016) and a modified NACA 0010-34 profile. Tests were performed at aspect ratios ($AR$) of 1.0 and 1.5, with Froude numbers ranging from 0.5 to 2.5.

The primary experimental approach involved an "alternative" surveying method: the hydrofoil was accelerated to a target Froude number at a fixed, shallow angle of attack, after which the angle of attack was gradually increased until the onset of ventilation. This was done under quasi-steady-state conditions. To ensure the validity of this assumption, a criterion was established based on potential flow theory (Tupper, 1983), requiring the number of convective timescales per degree of angle change ($\Delta\tau$) to exceed 20. This minimized unsteady hydrodynamic effects and surface disturbances.

Instrumentation included a three-axis force balance to measure hydrodynamic loads and a multi-camera imaging system (both above and below the waterline) to capture free-surface deformation and flow visualization. The study also included a limited set of tests following the conventional approach (varying $Fr$ at fixed $\alpha$) to validate previous data.

Key Results

1. Trigger Mechanisms: Three distinct mechanisms for the onset of ventilation were identified:

   • Nose Ventilation: Prevalent at lower Froude numbers ($Fr < 1.0$ for $AR=1.0$;$Fr < 1.25$ for $AR=1.5$). It is associated with the formation of a laminar separation bubble (LSB) and a subsequent rupture of the interfacial layer between the separated flow and the free surface. This process is relatively rapid (approx. $3.5\tau$).
   • Tail Ventilation: Prevalent at higher Froude numbers. Driven by Rayleigh–Taylor instability, surface perturbations grow into air-filled vortices under the downward acceleration field created by the hydrofoil. This is a more gradual process (approx. $7\tau$).
   • Base Ventilation: Observed only for the semi-ogive profile at high Froude numbers. It acts as a precursor involving wake vortices but did not lead to a stable ventilated cavity in the tested conditions.

2. Inception Angle of Attack: The study mapped the inception angle of attack ($\alpha_i$) as a function of $Fr$. A non-monotonic trend was observed: $\alpha_i$ increased with $Fr$ in the nose ventilation regime, peaked near the transition to tail ventilation, and then decreased as $Fr$ increased further. The peak $\alpha_i$ exceeded $25^\circ$, significantly higher than the $\sim 15^\circ$ boundary suggested by previous stability maps.

3. Discrepancy in Stability Maps: When the new data (varying $\alpha$ at constant $Fr$) was overlaid on the stability map proposed by Harwood et al. (2016) (derived by varying $Fr$ at constant $\alpha$), a stark discrepancy emerged. The current study found that ventilation onset was delayed to much higher angles of attack. The boundary between the globally stable fully wetted regime and the bistable regions is not uniform and extends to significantly higher $\alpha$ than previously estimated.

4. Hydrodynamic Coefficients: Measurements of lift and drag coefficients at the onset of ventilation showed that existing semi-empirical models (e.g., Damley-Strnad et al., 2019) align well with data in the tail ventilation regime but deviate significantly in the nose ventilation regime.

Significance and Claims
The paper claims that the assumption of path-independence in mapping ventilation stability is invalid. The method used to survey the parameter space (varying $\alpha$ vs. varying $Fr$) fundamentally alters the observed onset conditions. Specifically, the boundary between bistable and globally stable regions is not a fixed line at $\alpha \approx 15^\circ$ but is a complex function of the flow history and trigger mechanism.

The authors propose a revised stability map that reconciles the current data with previously published results. This map suggests that two alternative paths to a steady-state condition can lead to different flow regimes. For instance, increasing $Fr$ at a fixed $\alpha$ may trigger ventilation at lower angles than increasing $\alpha$ at a fixed $Fr$.

The study also derives a semi-empirical condition for tail ventilation, $Fr > AR^{-1/2}$, based on the instability of the free surface under downward acceleration. Finally, the authors suggest that these findings may not be universal to all hydrofoil geometries; thicker hydrofoils, which are less prone to the laminar separation bubble mechanism driving nose ventilation, might exhibit different stability boundaries, potentially allowing for higher operating envelopes. The work underscores the necessity of considering trigger mechanisms and surveying procedures when predicting ventilation onset for surface-piercing hydrofoils.