Passive transverse forcing of turbulent boundary-layer flow using sinusoidal surface grooves

This study experimentally demonstrates that passive transverse forcing of turbulent boundary-layer flow using sinusoidal surface grooves generates a converging-diverging flow pattern known as a Passive Stokes Layer, which offers limited frictional drag reduction potential that is likely negated by pressure drag and other losses.

The Gist

Imagine you are trying to slow down a river of air flowing over a surface, like the wind rushing over a car or an airplane wing. This fast-moving air creates "skin friction," a type of drag that wastes energy. Scientists have long known that if you could make the surface wiggle side-to-side (back and forth) very quickly, you could smooth out the turbulent air and save energy. However, building a surface that physically vibrates is like trying to build a car with a motorized, wiggling skin—it's too complex, expensive, and power-hungry.

This paper asks a simple question: Can we trick the air into thinking the surface is wiggling, just by carving a clever pattern into it?

The Idea: The "Meandering Road"

The researchers tried carving shallow, winding grooves into a flat surface, shaped like a sine wave (a smooth, rolling hill pattern). Think of it like drawing a winding river on a flat piece of paper.

Their hypothesis was based on a simple analogy: If you run down a winding path, your body naturally sways left and right to follow the curves. They hoped that air flowing over these winding grooves would be forced to sway side-to-side (spanwise) just like a runner on a track, creating the same "wiggling" effect as the active, vibrating surfaces, but without needing any motors.

What They Actually Found

Using high-speed cameras to watch the air flow (like a super-slow-motion movie), they discovered the reality was a bit more complex than their simple "runner on a track" idea.

1. The "Converging-Diverging" Dance: Instead of the air simply following the groove like a train on a track, the air did something more interesting. As the grooves curved, the air didn't just turn; it squeezed in and then fanned out.

   • Analogy: Imagine water flowing through a garden hose that has a wavy shape. Instead of just following the waves, the water squirts out sideways at the bends and then gets sucked back in. The air was doing a "converging-diverging" dance, creating a complex swirl pattern rather than a simple side-to-side slide.

2. The "Passive Stokes Layer": They found that this pattern created a special layer of air near the surface, which they called a "Passive Stokes Layer."

   • Analogy: Think of this as a two-layered blanket. The bottom layer (right against the surface) is sticky and slow (viscous), while the top layer is pushed by the shape of the grooves and moves faster (inertial). Together, they create a "wiggling" effect in the air, even though the surface itself is perfectly still.

3. The "Too Steep" Problem: They tested grooves of different depths and widths.

   • Analogy: If the grooves are too shallow, the air doesn't notice them. If they are just right, the air starts to sway effectively. But if the grooves get too steep (like a very sharp, jagged mountain path), the air gets confused and the "wiggling" effect stops getting stronger. It hits a ceiling.

Did It Save Energy?

This is the most important part. The researchers wanted to know if this "trick" actually reduced the drag (friction) enough to be useful.

• The Good News: The grooves did successfully create the side-to-side air movement needed to calm down the turbulence. They proved the mechanism works.
• The Bad News: While the air friction (skin drag) went down slightly, the shape of the grooves created a new problem: pressure drag.
  • Analogy: Imagine trying to push a flat board through water. It's hard. Now, imagine carving deep, winding canyons into that board. While the water might flow smoother along the sides, the canyons themselves create a "braking" effect, like a sail catching the wind. The energy saved by smoothing the flow was almost entirely canceled out by the extra resistance caused by the shape of the grooves.

The Bottom Line

The paper concludes that while these "winding road" grooves are a clever way to passively make air sway side-to-side, they are likely not a practical solution for saving energy in real-world applications.

The tiny amount of friction saved is probably wiped out by the extra drag caused by the grooves themselves. It's like trying to save money by buying a cheaper, lighter car, only to realize the new car has a giant parachute attached to the back that slows it down. The researchers suggest that while the physics is fascinating and the flow control works, the net result is likely a wash or even a loss in efficiency.

Technical Summary

Technical Summary: Passive Transverse Forcing of Turbulent Boundary-Layer Flow Using Sinusoidal Surface Grooves

Problem Statement
Active spanwise wall forcing is a proven strategy for reducing turbulent skin-friction drag, with reported reductions exceeding 45%. This technique induces a spatio-temporal wave of spanwise motion (a Stokes layer) that interacts with near-wall turbulence to suppress drag. However, active systems suffer from high power consumption, complex implementation, and significant net energy losses when accounting for the entire actuation chain. Consequently, there is a need for passive techniques that can induce a similar spanwise flow without external power. While previous passive methods, such as oblique wavy walls and shallow spherical dimples, have shown promise, they often fail to achieve net drag reduction due to pressure drag penalties. This study investigates an alternative passive surface geometry: sinusoidal undulations (SU). These consist of parallel, meandering streamwise grooves designed to induce an oscillatory spanwise component in the near-wall flow, analogous to active spatial Stokes layer (SSL) actuation, but driven by the static surface geometry rather than wall motion.

Methodology
The study employs experimental investigation combined with analytical modeling to characterize the flow over five distinct SU test surfaces in a turbulent boundary layer (TBL).

• Experimental Setup: Experiments were conducted in an open-return wind tunnel at two free-stream velocities ($U_\infty = 5$ and $7.5$ m/s), corresponding to friction Reynolds numbers ($Re_\tau$) of 1020 and 1410. The test surfaces were CNC-machined with varying groove amplitudes ($A$), depths ($d$), and widths ($D$), while maintaining specific scaling ratios ($A/\lambda_x$, $d/D$, $\lambda_z/D$).
• Diagnostics: Particle Image Velocimetry (PIV) was the primary diagnostic tool. Measurements were taken in two planes: a wall-parallel plane ($x, z$) at $y^+ \approx 15$ to capture spanwise flow structures, and a streamwise-wall-normal plane ($x, y$) along the groove centerline to resolve the vertical flow profile.
• Modeling: The flow mechanism was elucidated using a potential-flow model (inviscid, incompressible) to predict the outer flow solution driven by surface displacement. Additionally, a 3D viscous analytical model was derived to describe the full Passive Stokes Layer (PSL), accounting for both the inertial outer layer and the viscous inner layer required to satisfy the no-slip condition.

Key Contributions and Results

1. Flow Topology and Mechanism: Contrary to the initial hypothesis that the flow would simply follow the groove path, the experiments revealed a complex converging-diverging flow pattern. The flow is driven by a spanwise-periodic pressure gradient ($P_z$) induced by the surface geometry. This results in:

   • Primary Flow: Spanwise flow aligned with the grooves, strongest where the groove slope is steepest.
   • Secondary Flow: A cross-groove flow directed across the groove edges, creating a converging-diverging topology between grooves.
   • Passive Stokes Layer (PSL): The induced flow forms a two-region structure: a thin viscous inner layer near the wall and a thicker, inertial outer layer where the maximum spanwise velocity ($W_{max}$) occurs away from the surface.

2. Geometric Scaling: The study establishes a scaling law for the maximum spanwise velocity amplitude based on the potential-flow model:
   $$W_{pot} \propto U_c \frac{A}{\lambda_x} \frac{d}{D}$$
   where $U_c$ is the convection velocity. Experimental results confirm that the forcing efficiency is proportional to the groove depth-to-width ratio ($d/D$) and the groove amplitude-to-wavelength ratio ($A/\lambda_x$). The forcing efficiency is approximately 20% relative to an idealized model where flow perfectly follows the groove geometry.

3. Limits of Forcing: The spanwise velocity amplitude increases with groove amplitude up to a saturation point ($A/\lambda_x \approx 0.09$). Beyond this threshold, the surface slope becomes too steep, leading to flow separation and a breakdown of the linear potential-flow prediction. The valid scaling regime is identified as $A/\lambda_x \cdot d/D \lesssim 4.5 \times 10^{-3}$.

4. Turbulence Modification: The SU surfaces induce a reduction in near-wall turbulence levels, specifically a 13% reduction in the inner peak of the streamwise normal Reynolds stress for the largest amplitude case. However, the mean streamwise velocity profiles do not exhibit the momentum deficit characteristic of significant drag reduction.

5. Drag Reduction Potential: By matching the surface-averaged Stokes strain of the PSL to that of an active SSL, the authors estimate the equivalent active forcing amplitude. This comparison suggests that the PSL generates sufficient strain to theoretically reduce frictional drag by approximately 1.8% (extrapolated from active forcing data). However, preliminary direct-force measurements indicate that any skin-friction reduction is likely offset by pressure drag and other losses, resulting in at best a few percent of net drag reduction, or potentially a net drag increase.

Significance and Claims
The paper concludes that sinusoidal undulations are effective at passively generating a spanwise flow and a Passive Stokes Layer through an inertial, pressure-gradient-driven mechanism. The study successfully elucidates the flow physics, distinguishing between the inertial outer solution and the viscous inner solution, and provides a validated scaling law for the induced velocity.

Regarding practical application, the authors maintain a modest stance. While the passive forcing is sufficient to act on the spanwise-forcing drag-reduction mechanism, the net energy savings are likely negligible. The potential frictional drag reduction is small and is expected to be offset by pressure drag and implementation losses. Therefore, similar to other passive techniques like dimples and oblique wavy walls, the potential for net drag reduction in practical applications appears limited. The study serves primarily as a fundamental characterization of the flow physics and a quantitative assessment of the trade-offs between induced forcing and pressure losses.