Wavy-wall-based flow control for the suction side geometry of NACA4412 at Retau = 3000

This paper demonstrates that applying a wavy-wall geometry to the suction side of a NACA4412 airfoil at Re_tau = 3000 significantly delays turbulent separation and increases the friction coefficient by up to 42.3% by enhancing small-scale streamwise convection and sweeping motions, provided the geometry avoids inducing detrimental large-scale flow separations.

The Gist

The Big Idea: Smoothing the Rough Ride

Imagine you are driving a car on a highway. If the road is perfectly smooth, the car glides easily. But if the road has bumps, the car bounces, loses speed, and the engine has to work harder. In the world of aerodynamics (how air moves around objects like airplane wings or wind turbine blades), "bumps" in the air flow cause the air to separate from the surface. When air separates, it creates a chaotic wake behind the object, which acts like a giant parachute, slowing the object down and making it harder to lift.

This paper is about a clever trick to stop that air from separating. The researchers tried putting a wavy surface (like a gentle, rolling hill) on the bottom side of a wing (the "suction side") to see if it could keep the air glued to the surface longer.

The Experiment: The "Wavy Wall"

The team built a wind tunnel and tested a curved surface that looks like part of a wind turbine blade. They compared a smooth surface against a surface with a specific, carefully designed wavy pattern.

Think of the air flowing over the wing like a crowd of people running down a hallway.

• On the smooth wall: As the crowd runs, they get tired and start to stumble. Some people stop and turn around, creating a traffic jam (this is "separation").
• On the wavy wall: The researchers found that the waves act like a rhythmic drumbeat. They keep the runners (the air molecules) moving in a tight, efficient line. Instead of stumbling, the runners get a little push forward, keeping the crowd moving fast right up against the wall.

The Results: A Big Win

The results were surprisingly good. By adding these waves:

1. Friction Increased (in a good way): Usually, we want less friction. But here, the waves made the air "grip" the wall tighter. This increased friction by 42%.
2. Separation Delayed: Because the air was gripping the wall so well, it didn't peel away (separate) until much further down the wing.
3. Better Lift, Less Drag: The wing could generate more lift (upward force) and less drag (backward force). The researchers estimate this could increase the wing's lift by about 5%, which is a massive improvement for wind turbines and airplanes.

The Secret Sauce: Small Waves vs. Big Waves

The most interesting part of the paper is why this works. The researchers discovered that the size of the waves matters immensely.

The "Goldilocks" Zone:

• Too Small: Nothing happens.
• Just Right: The waves create tiny, chaotic swirls (small-scale turbulence) right next to the wall. Imagine a swarm of bees buzzing near the ground. These bees (small eddies) constantly mix the fast-moving air from above with the slow-moving air near the wall. This "mixing" acts like a conveyor belt, shoving high-energy air down to the surface to keep it moving. This is the sweeping motion mentioned in the paper.
• Too Big (The Danger Zone): If the waves are too tall or too long, they create giant, slow-moving swirls (large-scale turbulence). Imagine a giant whirlpool forming in a river. This actually slows the water down and causes the flow to break apart. The researchers found that if the wavy wall is too long, it creates these giant, lazy swirls that ruin the effect.

The "Stop Sign" Rule

The paper concludes with a very important rule for designing these wavy walls: You have to know when to stop.

Think of the wavy wall like a song. If you play the song for the perfect amount of time, it gets the crowd dancing. If you play it too long, the crowd gets bored and stops dancing.

• The researchers found that the wavy wall should stop exactly when the "giant swirls" (large-scale energy) start to take over from the "buzzing bees" (small-scale energy).
• If you keep the waves going past this point, the benefit disappears, and the wing actually performs worse than if it were smooth.

Why This Matters

This is a passive solution. That means it doesn't require motors, pumps, or electricity. You just mold the wing or blade into a wavy shape, and it works automatically.

For wind turbines, this means:

• They can spin in lower winds without stalling.
• They can generate more power.
• They are cheaper to run because there are no moving parts to break.

Summary

The researchers found that painting a gentle, rhythmic wave pattern on a wind turbine blade acts like a "traffic cop" for the air. It uses tiny, energetic swirls to keep the air moving fast against the surface, preventing it from peeling off. However, the pattern has to be perfectly tuned; if the waves are too big or too long, they create giant, lazy swirls that slow everything down. When done right, it's a simple, cheap way to make wind energy much more efficient.

Technical Summary

1. Problem Statement

Passive flow control strategies for aerofoils often aim to reduce skin friction by attenuating large-scale streaks, which can inadvertently delay the laminar-to-turbulent transition and cause earlier flow separation, reducing lift. Active methods (e.g., suction/blowing) are effective but face practical and economic constraints, particularly on wind turbine blades. Existing passive alternatives like vortex generators offer minimal benefits at large scales, while dimples or grooves lose effectiveness at high Reynolds numbers.

The specific challenge addressed in this study is turbulent boundary layer (TBL) separation on the suction side of a convex surface (mimicking a wind turbine blade) at high Reynolds numbers ($Re_\tau \approx 3000$). The goal is to delay separation and improve aerodynamic performance without the complexity of active control systems.

2. Methodology

• Experimental Setup: Experiments were conducted in a modular wind tunnel at Czestochowa University of Technology. The test section featured a 7.5m flat plate to develop a canonical turbulent boundary layer, followed by a curved surface section corresponding to the suction side of a NACA4412 airfoil at a 5° angle of attack.
• Flow Conditions: The inlet velocity was approximately $14.5 \, m/s$, corresponding to a friction Reynolds number of $Re_\tau \approx 3100$ (simulating a wind speed of $10 \, m/s$ on a large-scale turbine blade). The inlet turbulence intensity was kept below 0.45%.
• Wavy Wall (WW) Design:
  • A streamwise wavy wall section ($L=540$ mm, 4 periods) was installed on the convex surface.
  • Geometry: The amplitude $A(x)$ increased downstream to maintain a constant viscous-scaled amplitude of $A^+ = 170$. The period was $\lambda = 135$ mm, resulting in an effective slope of $\approx 0.15$.
  • Material: High-density polyurethane plate bent to follow the convex curvature.
• Instrumentation:
  • Hot-Wire Anemometry (HWA): Two probes were used: a standard 55P05 ($l=1.2$ mm) for the smooth surface and a modified 55P31 ($l=0.41$ mm) for the wavy wall to ensure $l^+ < 9$ and minimize spatial averaging effects in high-shear regions.
  • Measurements: Mean velocity, Reynolds stresses, skin friction coefficient ($C_f$), and boundary layer parameters ($\delta, \theta, H, \beta$) were measured at various streamwise locations.
• Comparison: Results were compared against an unmodified (smooth) NACA4412 surface and benchmarked against active suction techniques reported in literature.

3. Key Contributions

1. First High-$Re$ Convex Surface Application: This is the first experimental documentation of the wavy wall method applied to a convex surface resembling a wind turbine blade at high Reynolds numbers ($Re_\tau \approx 3100$).
2. Mechanism Clarification: The study provides new physical insights, identifying small-scale turbulent activity as the primary driver for the method's success. It establishes that the wavy wall enhances small-scale streamwise convection and sweeping motions, provided the geometry does not induce large-scale separation in the troughs.
3. Optimization Criterion: The paper refines the termination criterion for wavy walls. It argues that the limit is not a fixed value of the Rotta-Clauser pressure gradient parameter ($\beta$), but rather the emergence of a distinct outer peak in the streamwise Reynolds stress that exceeds the inner peak. Exceeding this point triggers detrimental large-scale instabilities.
4. Quantitative Performance Data: It quantifies the drag reduction and separation delay capabilities of the passive method, showing it rivals active suction techniques.

4. Key Results

• Skin Friction Enhancement: The wavy wall configuration increased the skin friction coefficient ($C_f$) by 42.3% in the integration area downstream of the wavy section (compared to a 13.4% local increase at the inlet).
• Separation Delay: The method caused a significant downstream displacement of the turbulent boundary layer separation point. The tangential shift was 8.3% of the chord length, which correlates to an estimated 5% increase in lift.
• Momentum and Boundary Layer Thickness:
  • The total momentum was preserved (momentum-loss thickness $\theta$ remained stable), indicating the method does not merely redistribute momentum but enhances transport.
  • The boundary layer thickness ($\delta$) was reduced by approximately 5% compared to the smooth surface, indicating a reduction in wall-normal convection.
• Turbulence Structure:
  • Small-Scale Dominance: The wavy wall enhanced small-scale energy near the wall, increasing the sweeping motion and momentum transport to the surface.
  • Large-Scale Suppression: When the wavy wall was extended too far (beyond $\beta \approx 8$), large-scale motions became dominant, causing flow separation in the troughs and increasing the momentum-loss thickness.
  • Reynolds Stress Profiles: The method flattened the outer maximum of the Reynolds stress profile and reduced the universal turbulence level near the wall, similar to the effects of active suction.
• Friction Reynolds Number: Unlike the smooth surface where $Re_\tau$ drops significantly downstream, the wavy wall maintained a nearly constant $Re_\tau$ along the flow, indicating sustained shear stress.

5. Significance

• Aerodynamic Efficiency: The study demonstrates that a carefully designed passive wavy wall can significantly improve the lift-to-drag ratio of wind turbine blades by delaying separation and reducing pressure drag, without the energy penalty of active systems.
• Design Guidelines: The research offers critical design rules for wavy walls:
  • Maintain high small-scale eddy density.
  • Ensure the effective slope keeps the flow in the troughs on the verge of separation but not separated.
  • Terminate the wavy wall before the outer peak of Reynolds stress surpasses the inner peak (typically around $\beta \approx 8$ for this flow history), rather than relying on a fixed $\beta$ limit.
• Physical Insight: It resolves discrepancies between previous numerical simulations (which showed large-scale energy increases) and experimental results by highlighting that excessive length or amplitude triggers large-scale instabilities that negate the benefits. The optimal design must maximize small-scale convection while suppressing large-scale separation.

In conclusion, the wavy wall acts as a sophisticated passive flow control mechanism that, when tuned to the specific flow history and Reynolds number, enhances momentum transport, delays separation, and improves aerodynamic performance comparable to active suction.