Fluid Dynamics · Aerodynamics · Physics
📅 Квітень 2026 ⏱ ≈ 13 хв читання 🎯 Intermediate–Advanced

Boundary Layer and Drag — Prandtl's Theory of Viscous Flow

For most of the 19th century, the Navier-Stokes equations were considered intractable for engineering flows. In 1904, Ludwig Prandtl proposed a revolutionary simplification: near any solid surface, the flow is dominated by viscosity in a thin boundary layer — and outside this layer, the fluid can be treated as inviscid (ideal). This insight unified the previously contradictory theories of ideal flow and viscous drag, and it remains the foundation of modern aerodynamics, hydrodynamics, and turbulence modeling.

1. Boundary Layer Equations

Prandtl's key observation: within a thin layer of thickness δ ≪ L (L = body length scale), the velocity changes rapidly from zero at the wall (no-slip condition) to the freestream value U∞. Inside the layer, viscous stresses dominate; outside, they are negligible. This separation justifies simplifying the full Navier-Stokes equations:

Full Navier-Stokes (steady, incompressible, 2D): ρ(u ∂u/∂x + v ∂u/∂y) = -∂p/∂x + μ(∂²u/∂x² + ∂²u/∂y²) ρ(u ∂v/∂x + v ∂v/∂y) = -∂p/∂y + μ(∂²v/∂x² + ∂²v/∂y²) Boundary layer simplification (δ ≪ L, Re = UL/ν ≫ 1): u ∂u/∂x + v ∂u/∂y = -1/ρ · dp/dx + ν ∂²u/∂y² [momentum] ∂p/∂y ≈ 0 [pressure constant across boundary layer] ∂u/∂x + ∂v/∂y = 0 [continuity] Boundary conditions: y = 0: u = 0, v = 0 (no-slip) y → ∞: u → U(x) (outer inviscid solution)

2. The Blasius Solution

For a flat plate at zero angle of attack, the outer pressure gradient dp/dx = 0. Blasius (1908) found a similarity solution using the variable η = y√(U/νx):

Similarity variable: η = y · √(U∞ / νx) Stream function: ψ = √(νxU∞) · f(η) ODE: 2f''' + f·f'' = 0 (Blasius equation) BCs: f(0) = f'(0) = 0, f'(∞) = 1 Results: δ₉₉ = 5.0 √(νx/U∞) = 5.0x / √Re_x (99% velocity thickness) τ_w = 0.332 ρ U∞² / √Re_x (wall shear stress) Cf = 0.664 / √Re_x (local skin friction coefficient) C_D = 1.328 / √Re_L (total drag coefficient on plate)

The boundary layer grows as δ ∝ √x — thickening downstream as slower fluid accumulates in the wake of faster-moving fluid above. For a flat plate at ReL = 10⁶, δ/L ≈ 5×10⁻³ — confirming the thin-layer assumption.

3. Laminar-Turbulent Transition

The laminar Blasius boundary layer is stable only below a critical Reynolds number. Small perturbations are amplified by the Tollmien-Schlichting (TS) instability — oscillatory waves that grow and eventually break down into turbulence.

Critical Re_x for transition: Re_x,crit ≈ 3.5 × 10⁵ (smooth plate, low turbulence) Fully turbulent onset: Re_x,trans ≈ 10⁶ Factors accelerating transition: - Surface roughness (trips the boundary layer) - Adverse pressure gradient (dp/dx > 0) - High freestream turbulence intensity - Convex surface curvature (centrifugal instability → Görtler vortices) Factors delaying transition: - Favorable pressure gradient (accelerating flow, dp/dx < 0) - Suction through porous surface - Smooth surface + low-turbulence environment

Aircraft designers work hard to delay transition — a laminar boundary layer has ~5× lower skin friction than a turbulent one. The laminar flow wing profiles on modern airliners maintain laminar flow over 30–50% of the chord, saving several percent of total cruise drag.

4. Turbulent Boundary Layer

Once turbulent, the boundary layer has a complex multilayer structure. The law of the wall describes the mean velocity profile in wall units u⁺ = u/u_τ, y⁺ = y·u_τ/ν (where u_τ = √(τ_w/ρ) is the friction velocity):

Viscous sublayer (y⁺ < 5): u⁺ = y⁺ Buffer layer (5 < y⁺ < 30): transition Log-law region (y⁺ > 30): u⁺ = (1/κ) ln y⁺ + B κ ≈ 0.41 (Kármán constant) B ≈ 5.1 Defect layer (outer region): u⁺ approaches freestream Turbulent flat plate skin friction (White correlation): Cf ≈ 0.074 / Re_x^(1/5) (up to Re ≈ 10⁷) Cf ≈ 0.455 / (log Re_L)²·⁵⁸ (Schlichting, wide Re range)

Turbulent boundary layers are 3–8× thicker than laminar layers at the same Reynolds number and produce much higher wall shear stress — but they are also far more resistant to separation, which dramatically affects pressure drag.

5. Flow Separation

Flow separation occurs when the boundary layer cannot remain attached to the surface under an adverse pressure gradient (dp/dx > 0 — pressure increasing in the flow direction, as on the rear of a cylinder or the upper surface of a stalled wing). The wall shear stress τ_w drops to zero and reverses — fluid near the wall flows backward.

Once separated, the flow forms a large low-pressure wake behind the body (large pressure drag) and may cause complete aerodynamic stall on lifting surfaces. Turbulent boundary layers resist separation better than laminar ones because higher momentum fluid is continuously mixed down to the wall by turbulent eddies.

Golf ball dimples: Golf balls have dimples specifically to trip the boundary layer to turbulence at low speeds (Re ~ 10⁵), delaying separation and reducing the large pressure drag that a laminar boundary layer would produce. This cuts drag by ~50% compared to a smooth ball at golf-ball speeds.

6. Types of Drag

Total drag = Skin friction drag + Pressure (form) drag + Induced drag + Wave drag Skin friction drag: D_f = ∫ τ_w dA [friction from boundary layer on surface] Dominant for streamlined bodies (flat plates, airfoils at small α) Pressure drag (form drag): D_p = ∫ p n̂·x̂ dA [pressure imbalance front vs. rear] Dominant for bluff bodies; caused by flow separation Induced drag (lift-induced, 3D only): C_Di = CL² / (π AR e) AR = aspect ratio, e = Oswald efficiency factor Minimum for elliptical lift distribution (e = 1) Wave drag: Occurs at transonic/supersonic speeds (M ≥ ~0.7) Energy carried away by shock waves

7. The Drag Crisis

The drag coefficient of a smooth sphere drops dramatically at a critical Reynolds number Re ≈ 4×10⁵ — from C_D ≈ 0.5 to C_D ≈ 0.1. This drag crisis is caused by the boundary layer transitioning from laminar to turbulent before the separation point:

The drag crisis affects soccer balls (Magnus effect becomes erratic), tennis balls (fuzz controls the transition Re), cricket balls (bowlers deliberately use one rough/one smooth hemispheres to exploit differential separation), and wind turbine aerodynamics. Understanding it requires detailed boundary layer analysis — no simple formula captures the non-monotonic C_D(Re) behavior.

💧 Try the Fluid Simulation →