Navier-Stokes equations and exact solutions (Couette, Poiseuille)

The incompressible Navier–Stokes equations, why they are hard, and their exact parallel-flow solutions: Couette, generalised Couette, plane Poiseuille and Hagen–Poiseuille pipe flow with wall shear and f = 64/Re.

Drafted with Aria, reviewed by the AiCanCode.org team. Spotted an error? Use Give Feedback at the bottom of the page.

Why it matters

The Navier–Stokes equations are the governing equations of viscous flow: every CFD code solves them, and boundary-layer theory, pipe-flow friction and lubrication theory all come from simplifying them. A few flows are simple enough to solve exactly. Couette and Poiseuille flows are the classic ones; they give the friction in bearings and seals, the laminar friction factor 64/Re, and benchmarks for checking numerical codes.

Key ideas

What the equations say. The Navier–Stokes (N–S) equations are Newton's second law for a fluid element with viscous stresses included: mass × acceleration (local plus convective) = pressure force + viscous force + body force. They are solved together with continuity (and, for compressible flow, the energy equation and an equation of state). For a Newtonian fluid the stress is linear in the strain rate (Stokes' hypothesis).

Incompressible, constant viscosity form. ρ DV/Dt = −∇p + μ∇²V + ρg, with ∇·V = 0. Dropping the viscous term gives Euler's equation. Four unknowns (u, v, w, p) and four equations.

Why they are hard. The convective term (V·∇)V is nonlinear; at high Reynolds number the solutions become unstable and turbulent. No general analytical solution exists. Exact solutions exist where the nonlinear term vanishes, typically fully developed parallel flows (velocity in one direction, varying only across the channel), so (V·∇)V = 0.

Boundary conditions. No-slip and no-penetration at solid walls (fluid velocity equals wall velocity); specified velocity or pressure at inlets and outlets; zero shear and constant pressure at a free surface.

Couette flow. Fluid between two parallel plates a gap h apart, the upper plate moving at U, no pressure gradient. The N–S x-equation reduces to d²u/dy² = 0, so u = U·y/h, a linear profile with uniform shear stress μU/h. Model for journal bearings with small clearance.

Generalised Couette flow. With a pressure gradient as well, d²u/dy² = (1/μ)dp/dx. The profile is the linear Couette part plus a parabola. An adverse gradient (dp/dx > 0) can produce backflow near the stationary wall; zero net flow occurs when dp/dx = 6μU/h².

Plane Poiseuille flow. Both plates fixed, gap b, flow driven by −dp/dx: u = (1/(2μ))(−dp/dx)·y(b − y), parabolic, u_max = 1.5 × mean.

Hagen–Poiseuille flow (pipe). Fully developed laminar flow in a pipe of radius R: u(r) = (1/(4μ))(−dp/dx)(R² − r²). u_max = 2 × mean velocity. Flow rate Q = πR⁴(−dp/dx)/(8μ) = πD⁴Δp/(128μL). Wall shear τ_w = 8μV̄/D, and the Darcy friction factor is f = 64/Re. Valid for Re below about 2300 and away from the entrance region.

Assumptions for all three. Steady, laminar, incompressible, Newtonian, fully developed (no change along the flow), gravity absorbed into the pressure or negligible.

Formulas

ρ·(∂V/∂t + (V·∇)V) = −∇p + μ·∇²V + ρ·g with ∇·V = 0

  • ρ (kg/m³); V (m/s); p (Pa); μ (Pa·s); g (m/s²). Incompressible Newtonian fluid, constant μ.

u = U·y / h, τ = μ·U / h

  • Couette flow; U: plate speed (m/s); h: gap (m); y measured from the fixed plate (m).

u = U·y/h + (1/(2μ))·(−dp/dx)·y·(h − y)

  • Generalised Couette flow; dp/dx in Pa/m.

u = (1/(2μ))·(−dp/dx)·y·(b − y), u_max = b²·(−dp/dx)/(8μ), q = b³·(−dp/dx)/(12μ)

  • Plane Poiseuille; b: gap (m); q: flow per unit width (m²/s).

u = (1/(4μ))·(−dp/dx)·(R² − r²), u_max = R²·(−dp/dx)/(4μ) = 2·V̄

  • Hagen–Poiseuille; R: pipe radius (m); r: radial position (m).

Q = π·D⁴·Δp / (128·μ·L), τ_w = 8·μ·V̄ / D, f = 64 / Re

  • Q (m³/s); D: diameter (m); Δp: pressure drop over length L (Pa); V̄: mean velocity (m/s).

Worked examples

Example 1 (standard): lubricating film. Given: μ = 0.1 Pa·s, plate speed U = 3 m/s, gap h = 0.6 mm.

  1. τ = μ·U/h = 0.1 × 3/0.0006 = 500 Pa.
  2. Velocity at mid-gap: u = U/2 = 1.5 m/s. Answer: τ = 500 Pa, uniform across the gap.

Example 2 (standard): laminar oil pipe. Given: oil μ = 0.3 Pa·s, ρ = 900 kg/m³, D = 20 mm, Q = 0.2 L/s.

  1. V̄ = Q/A = 0.0002/(π × 0.02²/4) = 0.6366 m/s.
  2. Re = ρV̄D/μ = 900 × 0.6366 × 0.02/0.3 = 38.2, laminar.
  3. Δp/L = 128·μ·Q/(π·D⁴) = 128 × 0.3 × 0.0002/(π × 1.6×10⁻⁷) = 15 279 Pa/m.
  4. τ_w = 8·μ·V̄/D = 8 × 0.3 × 0.6366/0.02 = 76.4 Pa. Check: τ_w = (Δp/L)·R/2 = 15 279 × 0.01/2 = 76.4 Pa.
  5. u_max = 2V̄ = 1.27 m/s. Answer: Δp/L ≈ 15.3 kPa/m, τ_w ≈ 76.4 Pa.

Example 3 (GATE level): generalised Couette flow with zero net flow. Given: upper plate speed U = 2 m/s, gap h = 2 mm, μ = 0.05 Pa·s. Find dp/dx for zero net flow.

  1. Flow per unit width: q = U·h/2 + h³·(−dp/dx)/(12μ).
  2. Set q = 0: dp/dx = 6μU/h² = 6 × 0.05 × 2/(0.002²) = 150 000 Pa/m.
  3. The gradient is adverse (pressure rises in the direction of plate motion); flow near the moving plate goes forward and flow near the fixed plate goes backward. Answer: dp/dx = 150 kPa/m.

Common mistakes

  • Using u_max = 2V̄ for flow between plates. That is for a pipe; between plates u_max = 1.5V̄.
  • Confusing radius and diameter in Q = πR⁴Δp/(8μL) versus πD⁴Δp/(128μL).
  • Losing the sign of dp/dx: flow goes from high to low pressure, so −dp/dx is positive for forward flow.
  • Applying Hagen–Poiseuille to turbulent flow (Re above about 2300) or to the entrance region.
  • Saying the N–S equations express mass and energy conservation. They are the momentum equations; continuity and energy are separate.

For GATE AE

Expect: simplifying N–S for a given parallel flow and integrating with boundary conditions; Couette shear stress and velocity at a point; maximum, mean velocity and flow rate in pipe and channel Poiseuille flow; wall shear stress and f = 64/Re; and the condition for flow reversal in generalised Couette flow. Practise writing the reduced equation, integrating twice and applying both boundary conditions cleanly.

Quick check

  1. In plane Poiseuille flow, what is u_max/V̄?
  2. Couette flow, μ = 0.002 Pa·s, U = 2 m/s, h = 5 mm. Find τ.
  3. What is the laminar Darcy friction factor at Re = 1600?
  4. Which term of the N–S equations vanishes in fully developed parallel flow?

Answers: 1. 1.5 2. 0.8 Pa 3. 0.04 4. The convective (nonlinear) acceleration term

Try answering each one aloud before you open it.

  1. 1.What are the Navier-Stokes equations and why are they important in fluid mechanics?Concept

    They are the momentum equations for a viscous (Newtonian) fluid: Newton's second law for a fluid element, with local and convective acceleration balanced by pressure, viscous and body forces. In incompressible form, ρ DV/Dt = −∇p + μ∇²V + ρg, solved together with continuity ∇·V = 0 (and an energy equation for compressible flow). They govern essentially all practical flows, so boundary-layer theory, pipe friction and every CFD code are built on them.

  2. 2.Explain the difference between Couette flow and Poiseuille flow.Concept

    Couette flow refers to the flow of a viscous fluid between two parallel surfaces, where one surface is moving relative to the other. It is characterized by a linear velocity profile. Poiseuille flow, on the other hand, occurs in a pipe or between two stationary parallel plates, driven by a pressure gradient, and has a parabolic velocity profile. Both are exact solutions of the Navier-Stokes equations under specific conditions.

  3. 3.How do boundary conditions affect the solutions of the Navier-Stokes equations?Concept

    Boundary conditions are crucial in solving the Navier-Stokes equations as they define how the fluid interacts with its surroundings. They can specify the velocity, pressure, or stress at the boundaries of the fluid domain. Properly defined boundary conditions ensure that the solution is physically realistic and unique. For example, no-slip conditions are often applied at solid boundaries, meaning the fluid velocity matches the velocity of the boundary.

  4. 4.Why is the Navier-Stokes equation considered challenging to solve analytically?Concept

    The Navier-Stokes equations are challenging to solve analytically due to their non-linear nature and the complexity of the interactions they describe. The equations involve multiple variables and partial derivatives, making them difficult to simplify. Additionally, the presence of turbulence in many practical fluid flows adds to the complexity, often requiring numerical methods and computational fluid dynamics (CFD) for solutions.

  5. 5.What happens to the velocity profile in a Couette flow if the moving plate suddenly stops?Application

    If the moving plate in a Couette flow suddenly stops, the velocity profile will gradually change from linear to zero velocity throughout the fluid. The fluid will experience a deceleration due to viscous forces, and over time, the entire fluid will come to rest. This process is governed by the diffusion of momentum through the fluid, which can be described by the unsteady Navier-Stokes equations.

  6. 6.Why is Poiseuille flow used as a first model for blood flow in small vessels, and where does it fail?Application

    In small vessels the Reynolds number is low, so the flow is laminar and pressure-driven through a roughly circular tube, which are the Hagen–Poiseuille conditions; it shows the strong R⁴ dependence of flow rate on vessel radius that explains why small changes in vessel diameter control blood flow. It fails because blood is non-Newtonian (shear-thinning), flow is pulsatile in arteries, vessels are elastic, and in true capillaries red cells pass in single file, so the continuum parabolic profile no longer applies.

  7. 7.Calculate the volumetric flow rate for laminar Poiseuille flow of water (μ = 0.001 Pa·s) in a tube of radius 1 mm under a pressure gradient of 10 Pa/m.Numerical

    Q = πR⁴(−dp/dx)/(8μ) = π × (0.001)⁴ × 10/(8 × 0.001) = 3.93×10⁻⁹ m³/s. The mean velocity is Q/(πR²) = 1.25 mm/s, so Re = ρV̄D/μ = 1000 × 0.00125 × 0.002/0.001 = 2.5, comfortably laminar, which confirms the formula applies.

  8. 8.What is the significance of the Reynolds number in the context of Navier-Stokes equations?Concept

    The Reynolds number is a dimensionless quantity that indicates whether a fluid flow is laminar or turbulent. It is defined as Re = (ρ·v·L) / μ, where ρ is the fluid density, v is the velocity, L is a characteristic length, and μ is the dynamic viscosity. In the context of Navier-Stokes equations, the Reynolds number helps determine the relative importance of inertial forces to viscous forces, influencing the flow regime and the complexity of the solution.

  9. 9.How does the presence of turbulence affect the solutions of the Navier-Stokes equations?Application

    Turbulence introduces additional complexity to the solutions of the Navier-Stokes equations due to its chaotic and unpredictable nature. It results in rapid fluctuations in velocity and pressure, making analytical solutions impractical. Instead, turbulence is often modeled using statistical methods or approximations like Reynolds-averaged Navier-Stokes (RANS) equations, which average the effects of turbulence over time.

  10. 10.Determine the shear stress at the wall for a Couette flow with a fluid viscosity of 0.002 Pa·s, a moving plate velocity of 2 m/s, and a gap of 0.005 m between the plates.Numerical

    The shear stress τ at the wall in a Couette flow is given by τ = μ·(du/dy), where μ is the dynamic viscosity and du/dy is the velocity gradient. For a linear velocity profile, du/dy = (velocity of moving plate) / (gap between plates). Substituting the given values: τ = 0.002·(2 / 0.005) = 0.8 Pa.

Finished this topic? Mark it so your progress, study plan and readiness keep up.

Stuck on something here?