Boundary Layer & Drag: Prandtl's Theory of Viscous Flow

Q: What was Prandtl's key insight about boundary layers?

In 1904, Ludwig Prandtl proposed that near any solid surface, flow is dominated by viscosity in a thin boundary layer of thickness δ much less than the body's length, while outside this layer the fluid can be treated as inviscid (ideal). This unified the previously contradictory theories of ideal flow and viscous drag and remains the foundation of modern aerodynamics.

Q: Why do golf balls have dimples?

Dimples trip the laminar boundary layer into turbulence at the low Reynolds numbers a golf ball flies at, delaying separation and shrinking the low-pressure wake behind the ball. This cuts drag by roughly 50% compared to a smooth ball at golf-ball speeds — a rough surface reducing drag is counterintuitive, but the pressure drag saved vastly outweighs the small skin-friction penalty.

Q: What causes the drag crisis on a sphere?

Around Reynolds number 4×10⁵, a smooth sphere's drag coefficient drops suddenly from about 0.5 to about 0.1. At low Re the laminar boundary layer separates near the equator, producing a large wake; at high Re it transitions to turbulent before separating, resisting separation until roughly 120° from the stagnation point, which shrinks the wake dramatically.

The Blasius solution: how a boundary layer grows

Prandtl's key observation: within a thin layer of thickness δ ≪ L, velocity changes rapidly from zero at the wall (no-slip) to the freestream value. Inside, viscous stresses dominate; outside, they're negligible — a simplification that makes the full Navier-Stokes equations tractable. For a flat plate at zero angle of attack, Blasius (1908) found an exact similarity solution:

δ₉₉ = 5.0 · x / √Re_x        (99% velocity thickness — grows as √x)
τ_w = 0.332 · ρU∞² / √Re_x    (wall shear stress)
C_f = 0.664 / √Re_x            (local skin friction coefficient)

The boundary layer thickens downstream as slower fluid accumulates in the wake of faster-moving fluid above it — for a flat plate at Re_L = 10⁶, δ/L is only about 5×10⁻³, confirming just how thin this "thin layer" really is relative to the body.

live demo · the boundary layer thickening along a plate● LIVE

Laminar-turbulent transition and flow separation

The laminar Blasius layer is stable only below a critical Reynolds number, around Re_x ≈ 3.5×10⁵ for a smooth plate — beyond it, the Tollmien-Schlichting instability grows into full turbulence. A laminar layer has roughly five times lower skin friction than a turbulent one, which is why aircraft designers work hard to delay transition, maintaining laminar flow over 30–50% of a wing's chord to save several percent of cruise drag.

Flow separation happens when the boundary layer cannot stay attached under an adverse pressure gradient — wall shear drops to zero and reverses, and a large low-pressure wake forms behind the body. Turbulent boundary layers resist separation far better than laminar ones, because higher-momentum fluid is continuously mixed down to the wall by turbulent eddies.

The drag crisis

The drag coefficient of a smooth sphere drops dramatically around Re ≈ 4×10⁵ — from C_D ≈ 0.5 to C_D ≈ 0.1 — because the boundary layer transitions from laminar to turbulent before separating. At low Re the laminar layer separates near the equator (about 80° from stagnation), leaving a large wake; at high Re the turbulent layer resists separation until roughly 120°, shrinking the wake dramatically. Golf ball dimples deliberately trigger this transition early, at the low Reynolds numbers a golf ball actually flies at — cutting drag by about 50% compared to a smooth ball at the same speed.

Frequently asked questions

What was Prandtl's key insight about boundary layers?

Near a surface, flow is dominated by viscosity in a thin boundary layer (δ ≪ L); outside it, the fluid can be treated as inviscid. This 1904 insight unified ideal flow and viscous drag theories.

Why do golf balls have dimples?

Dimples trip the boundary layer into turbulence early, delaying separation and shrinking the wake — cutting drag by roughly 50% compared to a smooth ball at golf speeds.

What causes the drag crisis on a sphere?

Around Re ≈ 4×10⁵, the boundary layer transitions from laminar to turbulent before separating, resisting separation to about 120° instead of 80° and shrinking the wake — Cd drops from ~0.5 to ~0.1.

Try it live

Everything above runs in your browser — open Blasius Boundary Layer and drag along a flat plate to watch the laminar boundary layer thicken and read wall shear and skin friction at each station.

▶ Open Blasius Boundary Layer simulation