greatitself.
Continue
advanced29 min studyLesson 179 in Physics

Boundary Layers and Flow Separation

Resolve thin viscous regions, estimate laminar wall shear, and distinguish steady separation from a universal zero-shear rule.

Before you begin
On this page

The question for today

What you’ll understand

  • Balance streamwise inertia against transverse momentum diffusion to obtain a boundary-layer scale.
  • Use local flat-plate correlations only under their laminar, attached, zero-pressure-gradient assumptions.
  • Interpret adverse pressure gradients and wall shear without extending a steady two-dimensional criterion to arbitrary flows.

A small viscosity does not imply a small viscous effect everywhere. Near a stationary solid, no slip forces fluid velocity to change from zero to an outer value across a short distance. The resulting gradients can keep viscous stresses important even when inertia dominates farther away. Boundary-layer theory separates these regions instead of deleting viscosity globally. Our goal is to estimate their scales and understand how an apparently thin region can determine a large wake, pressure drag, and loss of lift.

A thin region with a leading-order viscous term

Choose distance x along a smooth wall and y normal to it. For a steady, two-dimensional, incompressible Newtonian flow, let the streamwise velocity scale be U and the layer thickness be delta. Streamwise convection scales as U squared divided by x; transverse viscous diffusion scales as nu U divided by delta squared. Equating these gives delta divided by x of order the inverse square root of the local Reynolds number. A large Reynolds number permits a thin layer; it does not eliminate its dynamics.

Leading boundary-layer equations retain continuity, streamwise convection, the imposed outer pressure gradient, and the second y derivative of streamwise velocity. Pressure is approximately uniform across the thin layer and supplied by the outer flow. This reduction requires small thickness relative to streamwise variation and suitably gentle curvature. It is not uniformly valid at the leading-edge singularity or through substantial separated regions. A local Reynolds number also does not certify laminar flow: disturbances, roughness, and upstream history influence transition.

Rex=Uexν,δx∼Rex−1/2,u ∂xu+v ∂yu=−1ρdpedx+ν ∂y2u\mathrm{Re}_x=\frac{U_e x}{\nu},\qquad \frac{\delta}{x}\sim\mathrm{Re}_x^{-1/2},\qquad u\,\partial_xu+v\,\partial_yu=-\frac{1}{\rho}\frac{dp_e}{dx}+\nu\,\partial_y^2u
The scaling uses a characteristic edge velocity Ue. The displayed momentum equation assumes a steady, incompressible, thin attached layer with constant kinematic viscosity.↔ Scroll wide equations horizontally.

Wall shear and integrated momentum deficit

A velocity profile supplies more information than a single thickness. The displacement thickness integrates the fractional velocity deficit; the momentum thickness integrates that deficit weighted by local velocity. These quantities tell an outer-flow calculation how much mass and momentum transport differ from an ideal uniform stream. The conventional 99-percent thickness instead marks where velocity first reaches 0.99 of the edge value. None is interchangeable without specifying a profile, and their ratios can change as the flow approaches separation.

For a laminar flat plate in a uniform incompressible stream, the Blasius solution gives a local skin-friction coefficient of 0.664 divided by the square root of local Reynolds number. This coefficient normalizes wall shear by one half rho U squared. It is local, so integrating it from the leading edge produces a different average coefficient. A flat-plate estimate cannot be transferred unchanged to a strong pressure gradient, turbulent layer, heated compressible wall, or rough surface simply because all cases have the same Reynolds number.

τw=μ∂u∂y∣y=0,Cf,x=τw12ρUe2=0.664Rex,δ99≈5xRex\tau_w=\mu\left.\frac{\partial u}{\partial y}\right|_{y=0},\qquad C_{f,x}=\frac{\tau_w}{\tfrac12\rho U_e^2}=\frac{0.664}{\sqrt{\mathrm{Re}_x}},\qquad \delta_{99}\approx\frac{5x}{\sqrt{\mathrm{Re}_x}}
Wall shear uses the locally planar stationary-wall boundary-layer approximation. The coefficient and thickness estimate apply specifically to an attached Blasius laminar flat plate with zero streamwise pressure gradient.↔ Scroll wide equations horizontally.

Adverse pressure gradients and the limits of separation criteria

An increasing outer pressure in the downstream direction is adverse to forward motion. The outer inviscid balance then predicts a decreasing edge speed. Fluid already slowed near the wall is especially vulnerable to this deceleration, so the profile can flatten at the wall and develop near-wall reverse flow. For a smooth stationary wall in a steady two-dimensional setting, a downstream change from positive wall shear through zero to negative shear is the classical local indicator of separation. An isolated zero alone does not establish that full pattern.

Do not generalize this scalar test to unsteady or three-dimensional flow: separation there concerns transport geometry and can require a different diagnosis. Turbulent mixing can increase resistance to an adverse gradient while increasing skin friction, illustrating why lower friction need not mean lower total drag. A separate numerical calculation must also resolve steep gradients across the layer; refinement is evidence only when the relevant physical quantities converge.

Work it through

Estimate local shear on an attached laminar plate

Air flows at 4 m/s over a smooth flat plate. At x = 0.20 m, use rho = 1.2 kg/m³ and nu = 1.5 × 10⁻⁵ m²/s. Assume laminar attached flow with uniform edge speed. Estimate local Reynolds number, local wall shear, and 99-percent thickness; explain why these values cannot locate separation.

  1. Compute the local Reynolds number from the distance already traveled: Re_x = (4 × 0.20)/(1.5 × 10⁻⁵) = 53,333.3. Its square root is approximately 230.94. The thin-layer scaling is plausible, but laminarity remains an imposed condition rather than a conclusion from this number alone.

  2. Use the local Blasius coefficient: Cf,x = 0.664/230.94 = 0.002875. The dynamic pressure is (1/2)(1.2)(4²) = 9.6 Pa, so tau_w = 0.002875 × 9.6 = 0.02760 Pa. Keep the factor of one half in the normalization.

  3. Estimate delta_99 = 5(0.20)/230.94 = 0.004330 m, or 4.33 mm. Its ratio to x is about 0.0217. Positive wall shear is consistent with the assumed attached solution; a different pressure distribution would require a different profile calculation.

The result

The local estimates are Re_x ≈ 5.33 × 10⁴, tau_w ≈ 0.0276 Pa, and delta_99 ≈ 4.33 mm. They describe a zero-pressure-gradient solution and do not establish a separation location.

The living notebook

Explore the model

Interactive lab

Interactive lab · spectral

Spatial detail and near-wall resolution

The controls set mode count N and coefficient decay p in the finite sum Σ sin(nx−t)/nᵖ. Predict that increasing N at fixed p adds shorter spatial scales. Pause and compare a small and a large N, then increase p to suppress high modes.

Spatial detail and near-wall resolutionA finite sum of sin(nx − t)/nᵖ. Playback changes t; increasing the cutoff includes finer spatial scales. This is not a proof about a continuum limit or renormalization. Current readings: Sum at x = π/2: 0.867; Smallest wavelength: 2π / 5; Highest-mode coefficient: 0.2. All parameters are adjustable with the labeled controls below.01.573.144.716.28−2.56−1.2801.282.56x · radiansFinite sine sum
Finite sumHighest included mode
Sum at x = π/2
0.867
Smallest wavelength
2π / 5
Highest-mode coefficient
0.2
t = 0 s

A finite sum of sin(nx − t)/nᵖ. Playback changes t; increasing the cutoff includes finer spatial scales. This is not a proof about a continuum limit or renormalization. Parameter changes pause playback and recompute the model at the current elapsed time. Reset restores both parameters and returns time to zero. Playback starts paused. Adjust either parameter, then choose Play or Step.

Look forResolving fine spatial structure requires appropriate modes or grid spacing. This sine-sum illustration is not a boundary-layer profile, CFD calculation, or separation detector.

Pause & reason

Check your understanding

Which statement is justified when wall shear passes from positive to negative through zero along a smooth stationary wall?

Your turn

Try it yourself

Keep the worked-example fluid and edge speed, but move to x = 0.80 m while assuming the plate remains laminar and attached. Predict the changes in delta_99 and local wall shear before calculating. Does a smaller local shear prove that separation is imminent?

Take a moment to make a prediction or work it out on paper. Open a hint when you need one.

Give me a hint

Local Reynolds number grows by four. Its square root grows by two, while thickness contains both x and the inverse square root. The uniform outer speed still gives zero pressure gradient.

Compare with the solution

Re_x becomes 213,333.3 and its square root becomes 461.88. Therefore delta_99 = 5(0.80)/461.88 = 0.008660 m, twice the earlier thickness. Cf,x becomes 0.001438 and tau_w becomes 0.01380 Pa, half the earlier value. This positive shear decrease is ordinary downstream growth of an attached Blasius layer, not evidence of an adverse-gradient separation event. The numerical estimates remain conditional on laminarity; a real experiment could transition before reaching that station.

Keep the idea

Take it with you

  • High Reynolds number permits viscous effects to concentrate in thin regions with large gradients.
  • Local Blasius coefficients require their stated laminar flat-plate and zero-pressure-gradient assumptions.
  • Zero wall shear is a restricted classical separation indicator, not a universal rule for arbitrary flow.

Read further

Original sources and deeper explanations to follow at your own pace.

Optional progress, saved in this browser when available. Revisit any lesson at any time.

Lesson not marked complete.