7.5 Turbulent boundary layers

Turbulent vs Laminar Flow

Turbulent boundary layers are the thin, chaotic fluid regions that develop near solid surfaces at high speeds. They govern drag, heat transfer, and flow separation in nearly every engineering application, from aircraft wings to pipeline systems.

Before diving into turbulent boundary layers specifically, it helps to distinguish the two flow regimes:

• Turbulent flow features chaotic, irregular motion with rapid mixing and fluctuations in velocity, pressure, and temperature.
• Laminar flow features smooth, parallel layers of fluid with essentially no mixing between them.

The transition between these regimes depends on the Reynolds number. Low Reynolds numbers produce laminar flow; high Reynolds numbers produce turbulent flow. For flow over a flat plate, the critical Reynolds number is typically around $Re_x \approx 5 \times 10^5$, though surface roughness, freestream turbulence, and pressure gradients can shift this value.

Boundary Layer Theory

The boundary layer is the thin region of fluid near a solid surface where viscous effects are significant. It exists because of the no-slip condition: fluid velocity must equal zero at the surface. Moving away from the wall, velocity increases until it reaches the freestream value $U_\infty$.

Boundary Layer Thickness

Boundary layer thickness $\delta$ is defined as the distance from the surface where the local velocity reaches 99% of the freestream value. A few key points:

• $\delta$ increases with distance from the leading edge as more fluid is decelerated by viscous effects.
• Higher Reynolds numbers produce thinner boundary layers relative to the body length.
• Adverse pressure gradients cause the boundary layer to thicken more rapidly, while favorable gradients slow its growth.
• Surface roughness promotes earlier transition and can increase $\delta$.

Boundary Layer Separation

Separation occurs when the boundary layer detaches from the surface. The mechanism works like this:

1. An adverse pressure gradient (pressure increasing in the flow direction) decelerates the near-wall fluid.
2. The slowest fluid near the wall eventually stalls and reverses direction.
3. The reversed flow forces the boundary layer away from the surface, creating a separation bubble or fully separated wake.

Separation increases pressure drag, reduces lift (on airfoils), and introduces flow instability. Engineers delay or prevent it using flow control techniques like suction, blowing, or vortex generators.

Turbulent Boundary Layer Structure

A turbulent boundary layer is not uniformly chaotic. It has distinct regions, each with its own dominant physics and scaling laws. The key organizing concept is wall units, defined using the friction velocity $u_\tau = \sqrt{\tau_w / \rho}$, where $\tau_w$ is the wall shear stress and $\rho$ is the fluid density. Nondimensional wall-normal distance and velocity are:

$y^+ = \frac{y \, u_\tau}{\nu}, \qquad u^+ = \frac{u}{u_\tau}$

Inner vs Outer Region

• The inner region (close to the wall) is where viscous effects matter most. Velocity scales with wall units ($y^+$, $u^+$).
• The outer region (farther from the wall) is dominated by large-scale turbulent eddies. Velocity scales with the boundary layer thickness $\delta$ and the velocity defect $(U_\infty - u)$.
• An overlap region exists where both scaling laws hold simultaneously. This overlap is what gives rise to the logarithmic velocity profile.

Viscous Sublayer

This is the thinnest layer, right at the wall, where viscous shear stress overwhelms turbulent stresses. The velocity profile is linear:

$u^+ = y^+$

It extends from the wall to approximately $y^+ \approx 5$. Despite being extremely thin (often fractions of a millimeter), this layer controls the wall shear stress and heat transfer rate.

Buffer Layer

The buffer layer is the transition zone between viscous-dominated and turbulence-dominated behavior. It spans roughly $y^+ \approx 5$ to $y^+ \approx 30$. The velocity profile here doesn't follow either the linear law or the logarithmic law cleanly. This region is where turbulence production peaks, making it critical for drag and energy dissipation.

Logarithmic Layer

In this region the velocity profile follows the law of the wall:

$u^+ = \frac{1}{\kappa} \ln y^+ + B$

where $\kappa \approx 0.41$ is the von Kármán constant and $B \approx 5.2$ for smooth walls. The log layer extends from roughly $y^+ \approx 30$ to about $y/\delta \approx 0.2$. This is the most practically important region because it contains a large fraction of the boundary layer and its profile is remarkably universal across different flows.

Wake Region

Beyond the log layer, the velocity profile deviates upward toward the freestream value. This outer portion is often described using Coles' wake function, which adds a correction to the log law. The strength of the wake component depends on the pressure gradient and Reynolds number. Under adverse pressure gradients, the wake component grows significantly.

Turbulent Boundary Layer Equations

The governing equations come from the Navier-Stokes equations, but solving them directly (DNS) is computationally prohibitive for most engineering flows. Instead, engineers use averaged formulations.

Reynolds-Averaged Navier-Stokes (RANS) Equations

The RANS approach works in three steps:

1. Decompose every flow variable into a time-averaged mean and a fluctuating component: $u = \bar{u} + u'$.
2. Substitute into the Navier-Stokes equations and time-average the result.
3. The averaging process produces extra terms, the Reynolds stresses $-\rho \overline{u_i' u_j'}$, which represent momentum transport by turbulent fluctuations.

These Reynolds stress terms are the source of the closure problem.

Closure Problem

After averaging, the RANS equations contain more unknowns (the Reynolds stresses) than equations. You need a turbulence model to relate the Reynolds stresses to known mean-flow quantities. Common approaches include:

• Eddy viscosity models (e.g., Spalart-Allmaras, $k$-$\varepsilon$, $k$-$\omega$ SST): assume Reynolds stresses are proportional to mean strain rates through a turbulent viscosity $\nu_t$. Simpler and widely used, but struggle with flows involving strong separation or curvature.
• Reynolds stress models (RSM): solve transport equations for each Reynolds stress component. More accurate for complex flows but computationally expensive and harder to converge.

The right choice depends on the flow complexity and the accuracy you need.

Turbulent Boundary Layer Parameters

Reynolds Number Effects

As the Reynolds number increases:

• The boundary layer grows thicker in absolute terms, but thinner relative to the body length.
• The viscous sublayer and buffer layer shrink relative to $\delta$, meaning turbulent stresses dominate a larger fraction of the layer.
• Turbulent mixing intensifies, increasing both skin friction and heat transfer rates.
• The range of turbulent eddy scales widens, making the flow harder to simulate.

Pressure Gradient Effects

• A favorable pressure gradient ($dp/dx < 0$, i.e., accelerating flow) thins the boundary layer and stabilizes it against separation.
• An adverse pressure gradient ($dp/dx > 0$, i.e., decelerating flow) thickens the boundary layer and promotes separation.
• A zero pressure gradient produces the canonical flat-plate boundary layer, which is the baseline case for most correlations and turbulence model validation.

Surface Roughness Effects

Roughness elements on the surface affect the boundary layer in several ways:

• They can trigger earlier transition from laminar to turbulent flow.
• They increase skin friction drag by disrupting the viscous sublayer. Once roughness elements protrude beyond the sublayer ($k_s^+ > 5$, where $k_s$ is the equivalent sand-grain roughness height), the surface is considered hydraulically rough.
• They enhance heat transfer at the surface due to increased mixing.
• The effect depends on the size, shape, and spacing of roughness elements relative to the local boundary layer thickness and viscous length scale $\nu / u_\tau$.

Turbulent Boundary Layer Measurements

Experimental data are essential for validating turbulence models and understanding flow physics. Two dominant techniques are used.

Hot-Wire Anemometry

A hot-wire probe uses a thin electrically heated wire (typically a few micrometers in diameter) placed in the flow. As the flow velocity changes, convective cooling changes the wire's resistance, which is measured electronically.

• Strengths: Excellent temporal resolution (up to hundreds of kHz), capable of resolving turbulent fluctuations.
• Limitations: Poor spatial resolution if the wire is longer than the smallest turbulent scales. Sensitive to temperature drift, requiring careful calibration. Intrusive (the probe physically sits in the flow).

Particle Image Velocimetry (PIV)

PIV seeds the flow with small tracer particles, illuminates a plane with a laser sheet, and captures two successive images with a camera. Cross-correlating the images yields a velocity field.

• Strengths: Provides full 2D (or 3D with stereo/tomographic setups) velocity fields with high spatial resolution.
• Limitations: Temporal resolution is limited by camera frame rate (though time-resolved PIV systems are improving). Requires optical access and uniform particle seeding. Post-processing can be computationally intensive.

Turbulent Boundary Layer Control

Control techniques aim to modify the boundary layer to reduce drag, enhance heat transfer, or delay separation. They fall into two broad categories.

Passive vs Active Control

• Passive control relies on fixed geometric modifications (riblets, vortex generators, surface texturing). No energy input is needed, making these methods simple and robust, but they cannot adapt to changing flow conditions.
• Active control uses dynamic actuation (suction, blowing, synthetic jets, wall oscillation). These methods can respond in real time to flow changes but require an energy source and control system, so the net benefit must exceed the energy cost.

Riblets

Riblets are small streamwise grooves machined or applied onto a surface. They reduce turbulent skin friction by 5-8% under optimal conditions.

• They work by restricting the spanwise motion of near-wall streamwise vortices, which are the primary drivers of turbulent momentum transport near the wall.
• Optimal riblet spacing is roughly $s^+ \approx 15$ in wall units, meaning the physical size must be tailored to the local flow conditions.
• Shark skin inspired early riblet research; modern applications include aircraft fuselages and competitive swimsuits.

Polymer Additives

Adding small concentrations of long-chain polymers (e.g., polyethylene oxide) to a liquid flow can reduce turbulent drag by up to 80% in pipe flows. The polymers:

• Suppress near-wall turbulent fluctuations by absorbing energy from small-scale eddies.
• Effectively thicken the viscous sublayer, reducing the velocity gradient at the wall.
• Are most effective at low concentrations (tens of ppm), but the polymer chains degrade under shear over time, reducing effectiveness.
• Raise environmental concerns for open systems (e.g., marine applications).

Suction

Removing fluid through porous walls or discrete slots thins the boundary layer and re-energizes the near-wall flow. This:

1. Reduces the boundary layer thickness and displacement thickness.
2. Makes the velocity profile fuller (more momentum near the wall).
3. Delays or prevents separation under adverse pressure gradients.

The suction rate must be carefully controlled. Too little suction has negligible effect; too much can introduce flow disturbances or structural complications in the porous surface.

Turbulent Boundary Layer Applications

Aerodynamic Drag Reduction

Turbulent skin friction accounts for roughly 50% of total drag on a commercial aircraft. Reducing it through riblets, laminar flow control, or polymer coatings translates directly into fuel savings. Even a few percent reduction in skin friction drag on a long-haul aircraft can save millions of liters of fuel annually across a fleet.

Heat Transfer Enhancement

Turbulent boundary layers transfer heat much more effectively than laminar ones because of the intense mixing. Engineers exploit this in:

• Heat exchangers, where surface roughness or vortex generators break up the thermal boundary layer.
• Gas turbine blade cooling, where turbulent mixing helps protect blades from extreme combustion temperatures.
• Electronics cooling, where jet impingement creates thin, highly turbulent boundary layers for rapid heat removal.

The tradeoff is always increased drag or pressure drop versus improved thermal performance.

Flow-Induced Noise Reduction

Turbulent boundary layers generate broadband noise as pressure fluctuations on surfaces radiate as sound. This matters for:

• Aircraft (airframe noise during approach and landing)
• Wind turbines (trailing-edge noise is often the dominant noise source)
• Submarines (flow noise limits sonar performance)

Noise reduction techniques include trailing-edge serrations, porous surface treatments, and active flow control to weaken the turbulent pressure fluctuations near the surface.