4.4 Transition from laminar to turbulent flow

Laminar flow characteristics

Laminar flow is fluid motion where layers slide smoothly past one another with no cross-stream mixing. This is the "starting state" before transition occurs, so understanding its properties gives you the baseline for recognizing what changes during transition.

Low Reynolds numbers

The Reynolds number ($Re$) is a dimensionless ratio of inertial forces to viscous forces in a fluid. At low Reynolds numbers ($Re < 2300$ for pipe flow), viscous forces dominate. The fluid's viscosity is strong enough to damp out small disturbances before they can grow, keeping the flow smooth and orderly.

Smooth, parallel streamlines

In laminar flow, fluid particles travel along smooth, parallel paths called streamlines. These streamlines don't cross or intersect, which means adjacent fluid layers stay separate. The velocity profile is typically parabolic: maximum velocity at the center of the flow and zero velocity at the wall due to the no-slip condition.

Minimal mixing between layers

Because there are no turbulent eddies or vortices, laminar flow has very little mixing between fluid layers. This limited mixing means:

• Poor heat and mass transfer compared to turbulent flow
• Lower skin friction drag, since there's less momentum exchange between the fluid and the surface

Both of these characteristics become important when you compare laminar and turbulent boundary layers on a wing or inside a pipe.

Turbulent flow characteristics

Turbulent flow is the chaotic, irregular regime where eddies and vortices of many sizes drive rapid mixing and momentum transfer. It shows up at high Reynolds numbers, where inertial forces overpower viscous damping.

High Reynolds numbers

At high Reynolds numbers ($Re > 4000$ for pipe flow), the fluid's inertia is large enough to overcome viscous damping, so disturbances grow and spread rather than dying out. The specific Reynolds number at which transition occurs (the critical Reynolds number) depends on the flow geometry, surface conditions, and other factors.

Chaotic, irregular motion

Turbulent flow features fluid particles following complex, unpredictable paths. Eddies and vortices span a wide range of sizes, from large-scale structures down to tiny dissipative scales. Both velocity and pressure fluctuate rapidly in space and time.

Enhanced mixing and momentum transfer

The eddies in turbulent flow constantly stir the fluid, which produces:

• Much better heat and mass transfer than laminar flow
• Higher skin friction drag, because there's greater momentum exchange between the fluid and the surface

This trade-off between improved transfer and increased drag is central to many engineering design decisions.

Transition process

Transition is the sequence of events that takes a laminar boundary layer and turns it turbulent. It's not a sudden switch; it involves the growth, interaction, and eventual breakdown of flow instabilities.

Critical Reynolds number

The critical Reynolds number ($Re_{cr}$) marks the onset of transition. Its value depends on flow geometry, surface roughness, freestream turbulence, and pressure gradients. Some reference values:

• Pipe flow: $Re_{cr} \approx 2300$
• Flat plate (based on distance from leading edge): $Re_{cr} \approx 500{,}000$

These are approximate. In very quiet, smooth conditions, laminar flow on a flat plate can persist well beyond $Re = 500{,}000$.

Instability growth

Transition begins when small disturbances in the laminar flow start to amplify rather than decay. These disturbances can be triggered by:

• Surface roughness elements
• Freestream turbulence entering the boundary layer
• Adverse pressure gradients

The classic instability for a 2-D boundary layer is the Tollmien-Schlichting (T-S) wave, a sinusoidal oscillation of the velocity profile that grows exponentially once the local Reynolds number exceeds a threshold.

Nonlinear interactions

As the instabilities grow larger, they enter a nonlinear stage where they interact with each other and with the mean flow. This produces coherent structures such as hairpin vortices and streamwise streaks. Eventually, localized patches of turbulence called turbulent spots appear. These spots grow and merge until the entire boundary layer becomes fully turbulent.

The sequence, in summary:

1. Small disturbances appear in the laminar layer
2. Linear instabilities (e.g., T-S waves) amplify
3. Nonlinear interactions create 3-D structures and turbulent spots
4. Spots grow and merge into fully turbulent flow

Factors affecting transition

Several factors shift the critical Reynolds number up or down, making transition happen earlier or later. Knowing these is essential for both predicting and controlling transition.

Surface roughness

Roughness elements on a surface introduce disturbances directly into the boundary layer, acting as "trips" that force earlier transition. The effect depends on the height, shape, and spacing of the roughness relative to the local boundary layer thickness. A roughness element whose height is a significant fraction of $\delta$ will be far more effective at triggering transition than one buried deep in the sublayer.

Pressure gradient

• A favorable pressure gradient (pressure decreasing in the flow direction, as on the forward part of an airfoil) stabilizes the boundary layer and delays transition.
• An adverse pressure gradient (pressure increasing in the flow direction, as on the aft part of an airfoil) destabilizes the boundary layer and promotes earlier transition.

This is why airfoil designers shape the pressure distribution carefully: pushing the point of minimum pressure farther aft extends the laminar run and reduces drag.

Freestream turbulence

Higher turbulence intensity in the flow outside the boundary layer seeds disturbances into the layer, promoting earlier transition. The effect depends on both the intensity and the length scales of the freestream fluctuations relative to the boundary layer thickness. In wind tunnels, freestream turbulence levels are kept very low (often below 0.1%) to study "natural" transition; real flight environments are typically quieter than most tunnels.

Heat transfer

Temperature differences between the surface and the fluid alter the viscosity profile inside the boundary layer, which changes its stability:

• Cooling the surface (for gases) increases viscosity near the wall, which stabilizes the layer and delays transition.
• Heating the surface decreases near-wall viscosity, destabilizing the layer and promoting earlier transition.

The magnitude and distribution of the temperature gradient along the surface determine how strong this effect is.

Boundary layer theory

The boundary layer is the thin region near a surface where viscous effects matter and the velocity changes from zero at the wall to the freestream value. Boundary layer theory provides the mathematical framework for analyzing both laminar and turbulent layers.

Laminar boundary layer equations

The laminar boundary layer is governed by the Prandtl boundary layer equations, a simplified form of the Navier-Stokes equations. Key simplifications include assuming the layer is thin relative to the streamwise length scale and that the pressure across the layer is approximately constant (set by the outer inviscid flow). For simple geometries like a flat plate with zero pressure gradient, the Blasius solution gives an exact self-similar velocity profile.

Turbulent boundary layer equations

Turbulent boundary layers are far more complex because of the chaotic fluctuations. The standard approach is to decompose every quantity into a mean and a fluctuating part (Reynolds decomposition), then time-average the Navier-Stokes equations to get the Reynolds-Averaged Navier-Stokes (RANS) equations.

RANS introduces extra unknowns called Reynolds stresses (e.g., $-\rho \overline{u'v'}$), which represent the effect of turbulent fluctuations on the mean flow. Because there are more unknowns than equations, you need a turbulence closure model such as:

• $k\text{-}\varepsilon$ model
• $k\text{-}\omega$ model
• Spalart-Allmaras model

Each has strengths and weaknesses depending on the flow type.

Boundary layer thickness

The boundary layer thickness ($\delta$) is defined as the distance from the wall where the velocity reaches 99% of the freestream value. It grows with distance $x$ from the leading edge, but at different rates depending on the flow regime:

• Laminar: $\delta \propto x^{1/2}$ (slower growth)
• Turbulent: $\delta \propto x^{4/5}$ (faster growth)

The turbulent layer is thicker at the same $x$ location because turbulent mixing spreads momentum away from the wall more effectively.

Transition prediction methods

Predicting where transition occurs is critical for drag estimation, thermal design, and performance optimization. Methods range from simple correlations to full-resolution simulations.

Empirical correlations

These are data-driven formulas that relate the transition location to flow parameters like $Re$ and pressure gradient. They're fast and useful for preliminary design but limited to the range of conditions covered by the underlying experiments. Examples include the Michel criterion (flat plate) and the Abu-Ghannam and Shaw correlation (flows with pressure gradients).

Stability analysis

Stability analysis studies how small disturbances grow in a laminar boundary layer:

1. Linear Stability Theory (LST): Solve the Orr-Sommerfeld equation to find which disturbance frequencies are amplified and at what rate.
2. $e^N$ method: Integrate the growth rates of the most amplified disturbance along the surface. Transition is predicted when the total amplification factor reaches a critical value, typically $N \approx 9$ (though this varies with freestream turbulence level).
3. Parabolized Stability Equations (PSE): Capture nonlinear interactions between disturbances, giving more accurate predictions than LST alone, especially for 3-D flows.

Numerical simulations

High-fidelity simulations solve the Navier-Stokes equations directly or with minimal modeling:

• Direct Numerical Simulation (DNS): Resolves every scale of motion from the largest eddies down to the Kolmogorov dissipation scale. Extremely accurate but computationally expensive; currently limited to moderate Reynolds numbers.
• Large Eddy Simulation (LES): Resolves the large energy-carrying scales and models the small scales with a subgrid-scale model. A practical compromise between accuracy and cost.

Both DNS and LES provide detailed insight into the physics of transition, especially for complex geometries where empirical correlations and stability analysis fall short.

Transition control techniques

Engineers either want to delay transition (to reduce drag) or promote it (to prevent separation or enhance mixing). Control techniques fall into three broad categories.

Surface modifications

• Riblets: Small streamwise grooves that reduce near-wall turbulence intensity and can modestly delay transition or reduce turbulent skin friction.
• Superhydrophobic surfaces: Reduce surface friction by promoting slip at the wall, which can stabilize the boundary layer.
• Surface heating or cooling: Alters the viscosity profile near the wall to stabilize or destabilize the layer, as discussed above.

Active flow control

Active methods require external energy input:

• Boundary layer suction: Removes low-momentum fluid near the wall, thinning the boundary layer and stabilizing it. This is the basis of laminar flow control (LFC) on aircraft wings.
• Boundary layer blowing: Injects high-momentum fluid to energize the near-wall region, which can promote transition or prevent separation.
• Plasma actuators: Use electrical discharges to generate a body force in the fluid, offering fast-response, no-moving-parts control of the near-wall flow.

Passive flow control

Passive methods require no energy input:

• Vortex generators: Small vanes that create streamwise vortices, energizing the boundary layer to delay separation (though they can also promote transition).
• Turbulators: Roughness elements deliberately placed to trip the boundary layer turbulent, commonly used in heat exchangers to boost heat transfer.
• Porous surfaces: Allow transpiration of fluid through the wall, which can stabilize the boundary layer similarly to distributed suction.

Consequences of transition

Transition has major effects on drag, heat transfer, and noise. Whether those effects are beneficial or harmful depends entirely on the application.

Drag increase

A turbulent boundary layer has a much steeper velocity gradient at the wall than a laminar one, producing significantly higher skin friction drag. For an aircraft wing, turbulent skin friction can be 5 to 10 times greater than laminar skin friction at the same Reynolds number. This directly impacts fuel consumption and operating cost, which is why extending laminar flow is a major goal in aircraft design.

Heat transfer enhancement

The vigorous mixing in a turbulent boundary layer transports thermal energy away from (or toward) the surface much more effectively than laminar conduction alone. This is desirable in heat exchangers and turbine blade cooling, but it also means higher thermal loads on surfaces where you'd prefer less heat transfer.

Noise generation

Turbulent boundary layers produce broadband pressure fluctuations at the wall, which radiate as aerodynamic noise. This is a significant concern for aircraft cabin noise, wind turbine sound emissions, and automotive aeroacoustics. Maintaining laminar flow over noise-sensitive surfaces is one strategy for reducing aerodynamic noise.

Transition in various applications

Aircraft wings

On a wing, the boundary layer is typically laminar near the leading edge and transitions somewhere along the chord. Where that transition occurs determines the overall drag of the wing. Designers use natural laminar flow (NLF) airfoil shapes that maintain a favorable pressure gradient as far aft as possible, and in some cases add suction systems (hybrid laminar flow control) to push the transition point even farther back. Even small improvements in laminar run can yield measurable fuel savings on long-range aircraft.

Turbomachinery blades

Compressor and turbine blades operate at high speeds with strong pressure gradients, making transition prediction especially challenging. On the suction side of a turbine blade, the boundary layer may separate, transition while separated, and then reattach as a turbulent layer. Accurately predicting this separation-induced transition is critical for estimating blade losses and overall stage efficiency.

Pipe flows

In pipe flow, laminar conditions give a parabolic velocity profile and relatively low pressure drop (friction factor $f = 64/Re$). After transition, the velocity profile flattens and the pressure drop increases substantially. The critical Reynolds number is around 2300, but with very smooth entrance conditions and minimal disturbances, laminar flow has been maintained in experiments up to $Re \approx 100{,}000$. In practical piping systems, transition usually occurs near the classical value.

Atmospheric boundary layer

The atmospheric boundary layer (the lowest ~1-2 km of the atmosphere) is almost always turbulent during the day due to solar heating of the surface, but can become stably stratified and much less turbulent at night. Transition and turbulence in this layer govern the vertical transport of heat, moisture, and pollutants. Accurate modeling of these processes is essential for weather forecasting, air quality prediction, and wind energy resource assessment.

Experimental techniques

Experiments are essential for studying transition physics and validating computational models. The key challenge is capturing the rapid, small-scale velocity and pressure fluctuations that characterize the transition process.

Hot-wire anemometry

A hot-wire probe consists of a very thin wire (typically 5 µm diameter tungsten or platinum) heated by an electric current and placed in the flow. As the fluid cools the wire, its electrical resistance changes in proportion to the local velocity. Hot-wire anemometry offers excellent temporal resolution (response up to hundreds of kHz), making it well suited for measuring the high-frequency velocity fluctuations associated with instability growth and turbulent spots. Its main limitation is that it's an intrusive, point-measurement technique.

Particle image velocimetry

Particle Image Velocimetry (PIV) seeds the flow with small tracer particles and illuminates them with a laser sheet. A camera captures two successive images, and cross-correlation algorithms compute the displacement of particle groups between frames, yielding a 2-D (or stereo 3-D) velocity field. PIV provides whole-field, instantaneous snapshots of the flow, which is extremely valuable for visualizing the spatial structure of transitional flows. Its temporal resolution is lower than hot-wire anemometry, though high-speed PIV systems have narrowed this gap.

Flow visualization

Flow visualization techniques make the transition process directly visible:

• Smoke or dye injection reveals streamline patterns and the onset of turbulent mixing.
• Oil-film interferometry on a surface shows skin friction patterns, clearly marking where the flow transitions from laminar to turbulent.
• Infrared thermography exploits the difference in heat transfer between laminar and turbulent regions: the turbulent region transfers more heat, producing a distinct temperature signature on the surface.

These methods are especially useful for quickly identifying the transition location on complex geometries like swept wings or turbine blades.