The Setup: At extremely low Reynolds numbers ((Re \ll 1)), inertia is negligible, and the Navier-Stokes equations reduce to the linear Stokes equations. For a sphere of radius (a) moving with velocity (U) in a viscous fluid, Stokes derived the famous drag force (F = 6\pi\mu a U). However, this solution fails to satisfy the boundary conditions at infinity uniformly. In two dimensions, the Stokes paradox states no steady solution exists. In three dimensions, the Stokes solution is valid only as a leading-order approximation. The question: How do we find the first inertial correction to the drag?

The Solution (Oseen’s Approach): In 1910, Carl Wilhelm Oseen realized that far from the sphere, the inertial term (\rho (\mathbfu \cdot \nabla) \mathbfu) cannot be entirely neglected, even if (Re) is small. Instead, he linearized the inertia term around the uniform flow (\mathbfU):
[ (\mathbfu \cdot \nabla) \mathbfu \approx (\mathbfU \cdot \nabla) \mathbfu. ]
This yields the Oseen equations. Solving for flow past a sphere with matched asymptotic expansions (inner Stokes region near the sphere, outer Oseen region far away) gives the corrected drag:
[ F = 6\pi\mu a U \left[ 1 + \frac38 Re + O(Re^2 \ln Re) \right], \quad Re = \frac2\rho U a\mu. ]
The key insight: the (Re) correction comes from the long-range wake, which Stokes theory misses entirely. This problem teaches that singular perturbations—where a small parameter multiplies the highest derivative—require careful asymptotic matching.

Advanced fluid mechanics problems share common solution strategies:

The most challenging modern problems—turbulence closure, multiphase flows, non-Newtonian fluid dynamics—resist exact solutions. For these, the "solution" is a validated computational model that captures the dominant physics.

Whether you are preparing for a PhD qualifying exam or designing the next generation of hypersonic vehicles, mastering these advanced problems and their solutions transforms fluid mechanics from a collection of formulas into a powerful engineering intuition. Keep solving, keep questioning, and remember: in fluid dynamics, the flow is always more complex than it first appears.

Further Reading & References

Problem:
For steady laminar flow over a flat plate at zero incidence, use the Blasius similarity transformation ( \eta = y\sqrtU/(\nu x) ) and stream function ( \psi = \sqrt\nu U x f(\eta) ) to reduce the boundary layer equations to:
[ 2f''' + f f'' = 0 ] Boundary conditions: ( f(0)=0,\ f'(0)=0,\ f'(\infty)=1 ).
Given ( f''(0) \approx 0.332 ), compute the wall shear stress ( \tau_w ) and boundary layer thickness ( \delta_99 ).

Step 1: Determine the Flow Regime and Friction Factor Properties of water: $\nu \approx 1 \times 10^-6 , \textm^2/\texts$. Reynolds Number: $$ Re_D = \fracV D\nu = \frac4 \times 0.31 \times 10^-6 = 1,200,000 $$ The flow is turbulent ($Re > 4000$).

For turbulent flow in a smooth pipe ($4000 < Re < 10^5$), the Blasius correlation is appropriate: $$ f \approx \frac0.316Re_D^0.25 $$ $$ f \approx \frac0.316(1.2 \times 10^6)^0.25 = \frac0.31633.13 \approx 0.00954 $$

Step 2: Calculate Head Loss Using the Darcy-Weisbach equation: $$ h_f = f \fracLD \fracV^22g $$

For head loss per unit length ($h_f / L$): $$ \frach_fL = \fracfD \fracV^22g $$ $$ \frach_fL = \frac0.009540.3 \frac4^22(9.81) $$ $$ \frach_fL = 0.0318 \times \frac1619.62 = 0.0318 \times 0.8155 $$ $$ \frach_fL \approx 0.026 , \textm/m $$ (This represents a pressure drop of $\Delta P = \rho g h_f \approx 255 , \textPa$ per meter of pipe).

Step 3: Average vs. Max Velocity (1/7th Power Law) The turbulent velocity profile is approximated by: $$ u(r) = u_max \left( 1 - \fracrR \right)^1/7 $$ Where $r$ is the radial distance from the center and $R$ is the pipe radius.

To find the relationship between average velocity $V$ and $u_max$, we integrate over the pipe area $A = \pi R^2$: $$ V = \frac1\pi R^2 \int_0^R u_max \left(1 - \fracrR\right)^1/7 (2 \pi r) dr $$ Let $y = 1 - r/R$, so $r = R(1-y)$ and $dr = -R dy$. $$ V = \frac2 \pi R^2 u_max\pi R^2 \int_0^1 y^1/7 (1-y) dy $$ $$ V = 2 u_max \left[ \fracy^8/78/7 - \fracy^15/715/7 \right]0^1 $$ $$ V = 2 umax \left( \frac78 - \frac715 \right) = 2 u_max \left( \frac105 - 56120 \right) $$ $$ V = 2 u_max \left( \frac49120 \right) = u_max \left( \frac4960 \right) \approx 0.817 u_max $$

Result: $$ u_max = \fracV0.817 = \frac40.817 \approx 4.9 , \textm/s $$

Step 1: Simplify the Navier-Stokes Equations We start with the incompressible Navier-Stokes equation for the x-momentum: $$ \rho \left( \frac\partial u\partial t + u \frac\partial u\partial x + v \frac\partial u\partial y \right) = -\frac\partial P\partial x + \mu \left( \frac\partial^2 u\partial x^2 + \frac\partial^2 u\partial y^2 \right) $$

Given the assumptions:

The equation reduces to a simple balance between pressure and viscous forces: $$ 0 = -\fracdPdx + \mu \fracd^2 udy^2 $$ (Note: Partial derivatives become total derivatives as $u$ depends only on $y$.)

Step 2: Integrate the Differential Equation Rearranging gives: $$ \fracd^2 udy^2 = \frac1\mu \fracdPdx $$

Integrate once with respect to $y$: $$ \fracdudy = \frac1\mu \fracdPdx y + C_1 $$

Integrate a second time: $$ u(y) = \frac12\mu \fracdPdx y^2 + C_1 y + C_2 $$

Step 3: Apply Boundary Conditions

Step 4: Final Velocity Profile Substitute $C_1$ and $C_2$ back into the equation: $$ u(y) = \fracU yB - \frac12\mu \left(-\fracdPdx\right) (By - y^2) $$ (Here, we typically define a favorable pressure gradient as negative, so we swap signs for clarity).

Step 5: Condition for Zero Net Flow The flow rate per unit width is $Q = \int_0^B u(y) dy$. $$ Q = \int_0^B \left[ \fracU yB + \frac12\mu \fracdPdx (By - y^2) \right] dy $$ $$ Q = \fracU B2 + \frac12\mu \fracdPdx \left[ \fracB y^22 - \fracy^33 \right]_0^B $$ $$ Q = \fracUB2 + \frac12\mu \fracdPdx \left( \fracB^32 - \fracB^33 \right) $$ $$ Q = \fracUB2 + \fracB^312\mu \fracdPdx $$

For $Q = 0$: $$ \fracUB2 = - \fracB^312\mu \fracdPdx $$ $$ \fracdPdx = \frac6\mu UB^2 $$ This implies an adverse pressure gradient is required to exactly counteract the shear-driven flow from the moving plate.

The Problem: A viscous jet impinges normally on an infinite flat plate. The external potential flow is ( u_e = a x ), ( w_e = -2a z ) (axisymmetric). Determine the exact velocity profile.

The Advanced Solution Method: Use similarity transformation. For axisymmetric stagnation flow, the stream function ( \psi = r^2 f(z) ). The radial velocity ( u_r = (1/r) \partial\psi/\partial z = r f'(z) ). The vertical velocity ( u_z = -(1/r)\partial\psi/\partial r = -2 f(z) ).

Substituting into the Navier-Stokes equations reduces the PDE to an ODE (the axisymmetric Hiemenz equation): [ f''' + 2f f'' - (f')^2 + a^2 = 0 ] with boundary conditions: ( f(0)=0, f'(0)=0, f'(\infty)=a ).

This is solved numerically to find the wall shear stress ( \tau_w = \mu r f''(0) ). The value ( f''(0) \approx 1.312 ) is a universal constant.

Application: This solution models cooling of turbine blades by impinging jets and chemical vapor deposition reactors.