Boundary Layers and External Flow

Introduction and Context

When a real fluid flows past a solid surface, the no-slip condition requires the fluid velocity at the surface to equal the surface velocity. Viscous effects are therefore important near the surface, even when viscosity has little influence in the region far from it. The thin region in which velocity changes rapidly from the wall value to the external-flow value is the boundary layer.

Boundary-layer theory, introduced by Ludwig Prandtl in 1904, provides a practical bridge between inviscid-flow theory and real engineering flows. It explains skin-friction drag, pressure drag, flow separation, wake formation, lift generation, and performance losses in vehicles, turbines, pumps, aircraft, marine structures, and heat-transfer equipment.

Learning Objectives

By the end of this chapter, the reader should be able to:

1. Explain boundary-layer formation and distinguish laminar, transitional, and turbulent regimes.
2. Define displacement thickness, momentum thickness, and energy thickness.
3. Estimate boundary-layer thickness and wall shear stress for common external flows.
4. Explain adverse pressure gradients and flow separation.
5. Distinguish skin-friction drag from pressure or form drag.
6. Interpret lift and drag coefficients for bodies, wings, and hydrofoils.
7. Apply empirical correlations to engineering calculations and recognize their limitations.

1. Boundary-Layer Formation

Consider a uniform stream of velocity U∞ flowing over a stationary flat plate. At the leading edge, the fluid adjacent to the surface is brought to rest. Viscosity diffuses this velocity disturbance outward from the wall. The boundary layer consequently grows in the downstream direction.

The local boundary-layer thickness δ is conventionally defined as the distance from the wall at which the velocity reaches 99% of the external velocity:

Here u is the local streamwise velocity and y is the distance normal to the surface. This definition is conventional rather than exact; other thickness measures are often more useful for analysis.

The Reynolds number based on distance x from the leading edge is

where ρ is density in kg/m³, μ is dynamic viscosity in Pa·s, and ν = μ/ρ is kinematic viscosity in m²/s. Reynolds number compares inertial and viscous effects. Low values generally indicate strong viscous influence, while high values indicate a thin boundary layer and predominantly inertial external flow.

For a smooth flat plate, transition commonly begins around Reₓ ≈ 5 × 10⁵, although the actual value may range widely because of free-stream turbulence, surface roughness, pressure gradients, vibration, and acoustic disturbance.

Laminar and Turbulent Boundary Layers

In a laminar boundary layer, fluid moves in relatively orderly layers and momentum is transferred mainly by molecular viscosity. In a turbulent boundary layer, fluctuating three-dimensional motion produces additional momentum transfer, often represented through an effective or eddy viscosity. Turbulent layers are generally thicker and have greater wall shear stress, but they possess more near-wall momentum and therefore resist separation better than laminar layers.

For a smooth, zero-pressure-gradient flat plate, commonly used correlations are:

The turbulent expressions assume a smooth surface and are empirical engineering correlations. A turbulent flat-plate correlation sometimes used over a wider range is C̄_f = 0.074/Re_L⁰·² − 1742/Re_L, when the leading-edge laminar contribution is included. Correlations must not be extrapolated beyond their stated regime.

2. Integral Measures of Boundary-Layer Thickness

The velocity profile is more informative than a single geometric thickness. Three integral thicknesses are especially important.

Displacement Thickness

The displacement thickness represents the outward displacement of external streamlines caused by the velocity deficit:

The upper limit may be replaced by δ when the velocity is effectively U∞ beyond the boundary layer. A larger δ* indicates a stronger displacement effect on the outer inviscid flow.

Momentum Thickness

Momentum thickness measures the loss of streamwise momentum flux:

It is particularly useful in drag calculations and integral boundary-layer methods.

Energy Thickness

Energy thickness measures the kinetic-energy deficit:

The ratio H = δ*/θ is called the shape factor. For a typical laminar boundary layer, H is approximately 2.59 for a zero-pressure-gradient flat plate. A turbulent boundary layer commonly has H near 1.3–1.4. Increasing H generally indicates a fuller velocity deficit and a greater tendency toward separation.

3. Wall Shear Stress and Skin-Friction Drag

Newton’s law of viscosity gives the wall shear stress for a Newtonian fluid:

The local skin-friction coefficient is

Thus,

The reference dynamic pressure q∞ = ½ρU∞² has units of pascals. For a plate of wetted area A_w, the skin-friction drag is

For a two-dimensional plate of length L and width b, A_w = Lb for one wetted side. If both sides are exposed, the wetted area is approximately 2Lb, subject to the geometry and flow arrangement.

Worked Example 1: Flat-Plate Skin Friction

Air at 20 °C has ρ = 1.20 kg/m³ and ν = 1.50 × 10⁻⁵ m²/s. It flows at U∞ = 30 m/s over a smooth plate of length L = 2 m and width b = 1 m. Estimate the average skin-friction drag on one side, assuming turbulent flow over the plate.

First,

Using C̄_f = 0.074/Re_L⁰·²,

The dynamic pressure is

Since A_w = 2 m²,

The result is an estimate. Leading-edge shape, roughness, transition, and three-dimensional effects may change the actual drag.

4. Boundary-Layer Integral Analysis

The momentum-integral equation for a steady, incompressible, two-dimensional boundary layer may be written as

Here Uₑ(x) is the external velocity at the boundary-layer edge, θ is momentum thickness, H = δ*/θ, and τ_w is wall shear stress. The equation assumes a thin boundary layer, negligible streamwise viscous diffusion compared with normal diffusion, and known or estimable velocity-profile shape.

For zero pressure gradient, Uₑ is constant and the equation reduces to

This relation shows directly that momentum thickness grows because wall friction removes momentum from the flow.

5. Pressure Gradients and Flow Separation

The external pressure gradient strongly affects boundary-layer behavior. For inviscid outer flow, Bernoulli’s equation along a streamline gives

where elevation effects have been omitted for a horizontal flow. Therefore:

• If Uₑ increases downstream, pressure decreases: dp/dx < 0. This is a favorable pressure gradient.
• If Uₑ decreases downstream, pressure increases: dp/dx > 0. This is an adverse pressure gradient.

An adverse pressure gradient decelerates the boundary-layer fluid. Fluid close to the wall has low momentum and may be unable to move against the increasing pressure. The wall shear stress decreases and can become zero:

Downstream of separation, τ_w reverses sign and a recirculating region forms. The separated shear layer rolls into a wake, usually producing substantial pressure drag.

Separation is promoted by sharp turns, abrupt expansion, excessive airfoil incidence, surface roughness, and high adverse pressure gradients. It can be delayed by streamlining, gradual curvature, suction, blowing, vortex generators, or deliberate transition to turbulence.

6. Drag on External Bodies

The total drag on a body is the component of the surface-force resultant parallel to the free stream:

where S is the body surface, n_x is the streamwise component of the outward normal, and τ_xn represents the streamwise viscous traction. In practical engineering work, drag is expressed as

where A_ref is a selected reference area, usually frontal area for bluff bodies and planform area for wings.

Components of Drag

Skin-friction drag results from tangential viscous stresses. It is important for long, wetted, streamlined surfaces.

Pressure or form drag results from the pressure distribution around a body. It becomes large when separation creates a low-pressure wake behind the body.

Induced drag is associated with the three-dimensional generation of lift by finite wings. It arises because wing-tip vortices create downwash and tilt the aerodynamic force rearward.

Wave drag occurs in compressible flows when shock waves and other wave phenomena form, especially at transonic and supersonic speeds. It is outside the scope of incompressible boundary-layer theory but is important in high-speed aerodynamics.

Bluff and Streamlined Bodies

A circular cylinder or flat-faced object is a bluff body. Its separated wake is large, and pressure drag dominates. A streamlined body allows the flow to remain attached over a larger surface distance and therefore has lower pressure drag, although its wetted area may create greater skin friction.

For a circular cylinder, alternating vortex shedding can produce fluctuating lift and drag. The vortex-shedding frequency is often correlated using the Strouhal number:

where f is shedding frequency and D is cylinder diameter. For a broad range of subcritical Reynolds numbers, St is often approximately 0.2, but the value depends on Reynolds number, roughness, confinement, and free-stream turbulence.

7. Lift and Flow Around Aerofoils

The lift force is perpendicular to the free-stream direction and is commonly expressed as

For a wing, A_ref is normally planform area. Lift arises from the integrated pressure and shear distributions. In ordinary attached flow, changing angle of attack changes circulation and pressure difference between the two surfaces. A finite wing also produces induced downwash.

At small angles of attack, many aerofoils show an approximately linear relationship:

where C_{L0} is the lift coefficient at zero geometric angle, α is angle of attack in radians, and a is the lift-curve slope per radian. The relation fails near stall, where massive separation causes lift to reach a maximum and then decrease.

For a finite wing, an idealized induced-drag relation is

where AR = b²/S is aspect ratio, b is span, S is planform area, and e is the span-efficiency factor, normally less than or equal to 1. This equation assumes subsonic, attached, finite-wing flow and is most useful for preliminary estimates.

The total drag coefficient is often represented by a drag polar:

where C_{D0} is the zero-lift parasite-drag coefficient and k is approximately 1/(πeAR) when induced drag dominates the lift-dependent contribution.

8. Dimensional Analysis and Similarity

External-flow experiments are commonly organized using dimensionless groups. The most important are:

Dynamic similarity requires the relevant dimensionless groups to be comparable between model and prototype. For a ship model with a free surface, matching Froude number is often essential; matching Reynolds number simultaneously may be impossible. Engineers then use corrections and carefully selected test conditions.

9. Applied Case: Drag Reduction on a Vehicle Underbody

Background

This illustrative case concerns a small electric delivery vehicle travelling at 25 m/s. Its flat underbody and exposed suspension components create a separated wake and high aerodynamic drag. The design team considers a smooth underbody panel and gradual rear taper.

Problem

The vehicle has frontal area A = 2.4 m², air density ρ = 1.2 kg/m³, and an initial drag coefficient C_D = 0.38. The redesigned body is expected to reduce C_D to 0.30. Estimate the drag-force reduction.

Analysis

The drag equation is

The dynamic pressure is

Initial drag:

Redesigned drag:

Therefore,

At constant speed, the required aerodynamic power reduction is approximately

Approach and Outcome

The underbody panel reduces local separation caused by suspension components, while the taper decreases the pressure difference between the front and rear of the vehicle. Wind-tunnel testing or validated computational fluid dynamics should verify the predicted coefficient because ground clearance, rotating wheels, yaw angle, cooling flow, and crosswind can substantially alter the result.

Learning Points

1. Reducing pressure drag can be more valuable than reducing skin friction on a bluff vehicle.
2. A lower drag coefficient does not necessarily imply lower lift; vehicle stability must also be assessed.
3. Model-scale results require attention to Reynolds number, ground simulation, and wheel rotation.
4. Energy and environmental benefits should be quantified using realistic duty cycles rather than a single test speed.

10. Engineering Design, Measurement, and Computational Tools

Boundary layers are investigated using pressure taps, hot-wire anemometry, laser Doppler velocimetry, particle-image velocimetry, oil-flow visualization, force balances, and wake surveys. Computational fluid dynamics commonly uses Reynolds-averaged Navier–Stokes models, large-eddy simulation, or direct numerical simulation. Reynolds-averaged models are economical for design but depend strongly on turbulence-model assumptions, wall resolution, and separation treatment.

Best practice includes:

• define the reference area and reference velocity explicitly;
• report Reynolds number, surface roughness, turbulence intensity, and angle of attack;
• distinguish measured quantities from correlations;
• perform grid, time-step, and uncertainty checks in computational work;
• validate predictions against experiments or trusted benchmark data;
• check whether the flow is steady, transitional, separated, rotating, compressible, or free-surface dominated.

Artificial intelligence and machine learning can assist surrogate modelling, flow-field classification, reduced-order modelling, and design optimization. They do not remove the need for conservation laws, dimensional analysis, validation, or physical interpretability. Training data may be biased toward particular geometries or flow regimes, causing unsafe extrapolation.

11. Limitations and Professional Considerations

Boundary-layer correlations are approximations. They may fail for strongly three-dimensional flow, rough surfaces, strong curvature, massive separation, unsteady vortex shedding, compressibility, rotating machinery, or transitional flows. A formula derived for a smooth flat plate should not automatically be used for a vehicle, turbine blade, or ship hull.

Engineers must also consider noise generated by turbulent wakes, structural vibration from vortex shedding, cavitation for underwater bodies, emissions associated with energy consumption, and safety margins. In regulated applications, aerodynamic or hydrodynamic performance must be demonstrated through documented methods, traceable measurements, and applicable standards.

Chapter Summary

A boundary layer is the thin viscous region adjacent to a surface in which velocity changes from the wall value to the external-flow value. Reynolds number helps identify the relative importance of inertia and viscosity and provides a basis for estimating transition and boundary-layer thickness. Laminar layers have orderly motion and lower skin friction, whereas turbulent layers have higher momentum transfer, greater skin friction, and better resistance to separation.

Displacement thickness measures streamline displacement, momentum thickness measures momentum deficit, and energy thickness measures kinetic-energy deficit. Wall shear stress determines skin-friction drag. Adverse pressure gradients decelerate the boundary layer and may cause separation when wall shear stress falls to zero and then reverses. Separation produces wakes and pressure drag.

External-body performance is expressed through drag and lift coefficients. Bluff bodies are dominated by pressure drag, while streamlined bodies often experience a larger relative contribution from skin friction. Finite wings generate induced drag, which decreases as aspect ratio increases. Similarity parameters such as Reynolds, Strouhal, Froude, and Mach numbers guide experiments and model-scale interpretation. Reliable engineering decisions require appropriate correlations, validation, uncertainty assessment, and awareness of model limitations.

Key Terms

Boundary layer: Thin viscous region next to a solid surface.

External flow: Flow around an immersed body rather than through a closed passage.

Reynolds number: Dimensionless ratio of inertial to viscous effects.

Laminar flow: Relatively orderly flow with limited mixing between layers.

Turbulent flow: Unsteady, fluctuating flow with strong momentum mixing.

Transition: Change from laminar to turbulent boundary-layer behavior.

Displacement thickness: Equivalent outward displacement caused by the velocity deficit.

Momentum thickness: Integral measure of streamwise momentum loss.

Energy thickness: Integral measure of kinetic-energy loss.

Shape factor: Ratio δ*/θ used to characterize velocity-profile shape.

Wall shear stress: Viscous tangential stress acting at the wall.

Skin-friction drag: Drag caused by viscous surface shear.

Pressure drag: Drag caused by the pressure distribution, especially after separation.

Adverse pressure gradient: Pressure increasing in the flow direction.

Flow separation: Detachment of the boundary layer from the surface.

Wake: Region downstream containing velocity deficit and often vortices.

Lift coefficient: Dimensionless lift based on dynamic pressure and reference area.

Induced drag: Drag associated with finite-wing lift and downwash.

Strouhal number: Dimensionless frequency describing unsteady shedding.

Dynamic similarity: Matching of relevant dimensionless parameters between flows.

Review Questions

Short Answer Questions

1. What physical condition produces a boundary layer on a stationary wall?
2. Define displacement thickness and momentum thickness.
3. Why does an adverse pressure gradient promote separation?
4. Distinguish skin-friction drag from pressure drag.
5. What information is represented by the Reynolds number?

Long Answer Questions

1. Explain the development of a boundary layer over a flat plate, including transition and the differences between laminar and turbulent flow.
2. Derive or explain the physical meaning of the momentum-integral equation for a zero-pressure-gradient boundary layer.
3. Discuss the mechanisms of flow separation and methods used to delay it.
4. Explain how lift and induced drag are generated by a finite wing and state the assumptions behind the induced-drag relation.
5. Compare experimental, empirical, and computational approaches for predicting external-flow drag.

Analytical / Application Questions

1. Water with ρ = 998 kg/m³ and ν = 1.00 × 10⁻⁶ m²/s flows at 2 m/s over a smooth flat plate of length 0.5 m. Calculate Re_L and determine whether the plate is likely to experience transition before its trailing edge using Re_critical ≈ 5 × 10⁵.
2. A body experiences drag coefficient 0.32 at 20 m/s and 0.28 after redesign. If ρ = 1.2 kg/m³ and the reference area is 3 m², calculate the drag reduction.
3. Explain why triggering turbulence with a small surface device can sometimes reduce total drag on a body, even though turbulent skin friction is higher than laminar skin friction.

Multiple Choice Questions with Answers

Question 1

Which condition defines the conventional boundary-layer thickness δ?

A. u = 0.50U∞

B. u = 0.90U∞

C. u = 0.99U∞

D. u = 1.50U∞

Correct Answer: C. u = 0.99U∞

Question 2

For a smooth flat plate, which expression gives the local laminar skin-friction coefficient?

A. C_f,x = 0.664/√Reₓ

B. C_f,x = 0.0592/Reₓ⁰·²

C. C_f,x = Reₓ/0.664

D. C_f,x = 1/(2Reₓ)

Correct Answer: A. C_f,x = 0.664/√Reₓ

Question 3

At incipient boundary-layer separation, the wall shear stress is approximately:

A. Equal to dynamic pressure

B. Zero

C. At its maximum positive value

D. Equal to atmospheric pressure

Correct Answer: B. Zero

Question 4

An adverse pressure gradient means that pressure:

A. Decreases in the flow direction

B. Remains constant in the flow direction

C. Increases in the flow direction

D. Oscillates only in the wall-normal direction

Correct Answer: C. Increases in the flow direction

Question 5

What does momentum thickness primarily measure?

A. Loss of streamwise momentum flux

B. Increase in fluid temperature

C. Wave speed in compressible flow

D. Vortex-shedding frequency

Correct Answer: A. Loss of streamwise momentum flux

Question 6

Air with ρ = 1.2 kg/m³ flows at 10 m/s over an area of 2 m². If C_D = 0.5, what is the drag?

A. 60 N

B. 100 N

C. 120 N

D. 240 N

Correct Answer: C. 120 N

Question 7

Which body generally has pressure drag as its dominant drag component?

A. A highly streamlined fairing

B. A smooth long flat plate aligned with the flow

C. A bluff cylinder normal to the flow

D. An ideal inviscid streamline

Correct Answer: C. A bluff cylinder normal to the flow

Question 8

For a finite wing, induced-drag coefficient is approximately proportional to:

A. C_L²/(πeAR)

B. πeAR/C_L²

C. C_L/(ρU∞)

D. Re_L²/C_L

Correct Answer: A. C_L²/(πeAR)

Question 9

The Strouhal number is defined as:

A. ρUL/μ

B. fL/U

C. U/√(gL)

D. U/a

Correct Answer: B. fL/U

Question 10

Why can a turbulent boundary layer resist separation better than a laminar one?

A. It has no wall shear stress

B. It always has lower drag

C. It transports higher-momentum fluid toward the wall

D. It eliminates adverse pressure gradients

Correct Answer: C. It transports higher-momentum fluid toward the wall