Navier-Stokes equations and exact solutions

The incompressible Navier-Stokes equations term by term, and how to reduce them to exact solutions for Couette, plane Poiseuille, pipe and falling-film flows.

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 complete momentum balance for a Newtonian fluid; every CFD package solves them. A handful of flows — between plates, in pipes, in thin falling films — can be solved exactly, and these exact solutions are the backbone of lubrication, falling-film absorbers and evaporators, coating flows, viscometers and laminar heat- and mass-transfer analysis. They also show how to reduce a general equation to a problem you can integrate by hand.

Key ideas

What the equations say. For a fluid element: mass × acceleration = pressure force + viscous force + body force. Together with continuity they give four equations for u, v, w and p (for incompressible flow).

Assumptions behind the usual (incompressible) form.

  • Continuum fluid.
  • Newtonian: viscous stress is linear in strain rate (Stokes' constitutive relations).
  • Constant density and constant viscosity. The general compressible form also exists; it adds a ∇(∇·V) term and is used for gas dynamics. Temperature-dependent properties need the energy equation as well.

Meaning of each term (x-direction): ρ(∂u/∂t + u·∂u/∂x + v·∂u/∂y + w·∂u/∂z) is inertia (local + convective acceleration); −∂p/∂x is the pressure force; μ∇²u is the net viscous force; ρg_x is gravity. Dropping the viscous term gives Euler's equation; dropping the inertia term gives creeping (Stokes) flow, valid for Re ≪ 1.

Why exact solutions are rare. The convective terms are non-linear, so solutions cannot be superposed. Exact solutions exist where the geometry makes the convective terms vanish identically: parallel flows, in which only one velocity component exists and it varies only across the flow (fully developed).

Recipe for an exact solution.

  1. Choose coordinates aligned with the flow (Cartesian for plates and films, cylindrical for pipes and annuli).
  2. State assumptions: steady, fully developed, one velocity component, no variation in the third direction.
  3. Use continuity to show that the convective terms vanish.
  4. Reduce the momentum equation to an ordinary differential equation and integrate twice.
  5. Apply boundary conditions: no slip at a solid wall (fluid velocity equals wall velocity); zero shear (du/dy = 0) at a free surface in contact with a gas; symmetry (du/dy = 0) on a centreline.

Classical exact solutions.

  • Plane Couette flow: fluid between a fixed and a moving plate, no pressure gradient: linear profile u = U·y/h.
  • Plane Poiseuille flow: fixed plates, pressure-driven: parabolic profile, u_max = 1.5 × average.
  • Couette–Poiseuille: moving plate plus pressure gradient; the profiles superpose because the reduced equation is linear.
  • Hagen–Poiseuille: circular pipe, parabolic in r, u_max = 2 × average.
  • Falling film on an inclined or vertical plate: gravity-driven, half-parabola with the maximum at the free surface — the basis of Nusselt's film theory for wetted-wall columns and condensation.
  • Others: flow in an annulus, flow between rotating cylinders (Couette viscometer), Stokes' first problem (suddenly started plate).

Formulas

  • ρ·(∂u/∂t + u·∂u/∂x + v·∂u/∂y + w·∂u/∂z) = −∂p/∂x + μ·(∂²u/∂x² + ∂²u/∂y² + ∂²u/∂z²) + ρ·g_x — x-momentum, incompressible Newtonian; u, v, w (m/s), p (Pa), μ (Pa·s), ρ (kg/m³).
  • ∂u/∂x + ∂v/∂y + ∂w/∂z = 0 — continuity.
  • u = U·y/h — plane Couette flow; U plate speed (m/s), h gap (m).
  • u = (1/2μ)·(−dp/dx)·(h² − y²) — plane Poiseuille flow between plates at y = ±h (half-gap h).
  • u_max = (−dp/dx)·h²/(2μ), u_avg = (2/3)·u_max, q = (−dp/dx)·(2h)³/(12μ) — q flow per unit width (m²/s).
  • τ_w = (−dp/dx)·h — wall shear stress between plates (Pa).
  • u = (ρ·g·sinθ/μ)·(δ·y − y²/2) — falling film on a plane inclined at θ to the horizontal; y from the wall, δ film thickness (m).
  • u_max = ρ·g·sinθ·δ²/(2μ), u_avg = ρ·g·sinθ·δ²/(3μ), q = ρ·g·sinθ·δ³/(3μ) — film results; vertical wall θ = 90°.
  • Re_film = 4·ρ·q/μ — film Reynolds number; smooth laminar only up to about 30, wavy-laminar to about 1000–1800 (varies by source), then turbulent.

Worked examples

Example 1 (standard). Oil (μ = 0.3 Pa·s) flows between two fixed parallel plates 10 mm apart under a pressure gradient of −2000 Pa/m. Find u_max, the flow per metre width and the wall shear stress.

  1. Half-gap h = 0.005 m.
  2. u_max = (−dp/dx)·h²/(2μ) = 2000 × 2.5 × 10⁻⁵ / 0.6 = 0.0833 m/s.
  3. q = (−dp/dx)·(2h)³/(12μ) = 2000 × (0.01)³ / 3.6 = 5.56 × 10⁻⁴ m²/s. Check: u_avg × gap = (2/3 × 0.0833) × 0.01 = 5.56 × 10⁻⁴ ✓.
  4. τ_w = (−dp/dx)·h = 2000 × 0.005 = 10 Pa. u_max ≈ 0.083 m/s, q ≈ 5.56 × 10⁻⁴ m³/s per metre width, τ_w = 10 Pa.

Example 2 (GATE level). Water (ρ = 1000 kg/m³, μ = 1 × 10⁻³ Pa·s) runs down a vertical wall as a laminar film 0.3 mm thick. Derive and evaluate u_max, u_avg, the flow per metre of wall width and the film Reynolds number.

  1. Assumptions: steady, fully developed, u = u(y) only, no pressure gradient along the film (free surface at atmospheric pressure).
  2. x-momentum reduces to μ·d²u/dy² = −ρg.
  3. Integrate: du/dy = −(ρg/μ)y + C₁. Zero shear at the free surface y = δ gives C₁ = ρgδ/μ.
  4. Integrate again with u = 0 at y = 0: u = (ρg/μ)(δy − y²/2).
  5. u_max (at y = δ) = ρgδ²/(2μ) = 1000 × 9.81 × (3 × 10⁻⁴)² / (2 × 10⁻³) = 0.441 m/s.
  6. u_avg = (2/3)u_max = 0.294 m/s; q = u_avg·δ = ρgδ³/(3μ) = 8.83 × 10⁻⁵ m²/s.
  7. Re_film = 4ρq/μ = 4 × 1000 × 8.83 × 10⁻⁵ / 10⁻³ = 353 — in the wavy-laminar range, so the smooth-film result is an approximation. u_max ≈ 0.441 m/s, u_avg ≈ 0.294 m/s, q ≈ 8.83 × 10⁻⁵ m³/s per metre, Re_film ≈ 353.

Common mistakes

  • Using the full gap where the half-gap h appears (factor of 4 in u_max).
  • Applying no-slip at a free surface; a gas–liquid interface has zero shear, not zero velocity.
  • Forgetting that −dp/dx is positive for flow in +x.
  • Writing u_max = 2u_avg for plates (that is the pipe result; plates give 1.5).
  • Assuming exact solutions apply at high Re; they require laminar, fully developed flow.

For GATE CH

Expect derivation-based numericals for flow between plates, Couette–Poiseuille combinations, falling films on inclined plates, flow in an annulus, and questions identifying which NS terms vanish under given assumptions. Practise integrating d²u/dy² = constant and applying no-slip and zero-shear conditions cleanly.

Quick check

  1. Which term of the NS equations makes them non-linear?
  2. What boundary condition applies at the free surface of a falling film?
  3. Ratio u_max/u_avg for flow between fixed parallel plates?
  4. If film thickness doubles, by what factor does the flow per unit width change?

Answers: 1. The convective acceleration term (V·∇)V. 2. Zero shear stress, du/dy = 0. 3. 1.5. 4. 8 times (q ∝ δ³).

Try answering each one aloud before you open it.

  1. 1.What are the Navier-Stokes equations and what do they describe in fluid mechanics?Concept

    The Navier-Stokes equations are a set of partial differential equations that describe the motion of viscous fluid substances. They express the conservation of momentum and mass for fluid flow. These equations account for the forces acting on a fluid element, including pressure, viscous, and external forces, and are fundamental in predicting how fluids flow under various conditions.

  2. 2.Explain the significance of the continuity equation in the context of the Navier-Stokes equations.Concept

    The Navier–Stokes equations give three momentum equations, but there are four unknowns in incompressible flow (u, v, w and p); continuity, the mass balance, closes the set. For constant density it reduces to ∇·V = 0: the constant density is the assumption, and zero divergence is the consequence. In exact solutions continuity is also the step that tells you the convective terms vanish, for example in fully developed flow ∂u/∂x = 0 follows directly from continuity when v = w = 0.

  3. 3.What are the assumptions made in deriving the Navier-Stokes equations?Concept

    The fluid is treated as a continuum, and it is Newtonian, so viscous stress is linear in the rate of strain (with Stokes' hypothesis relating the bulk viscosity terms). Momentum conservation is applied to a fluid element with pressure, viscous and body forces. The general form allows compressible flow and variable properties; the familiar textbook form ρDV/Dt = −∇p + μ∇²V + ρg further assumes constant density and constant viscosity. They are not restricted to laminar flow — turbulence is contained in them — but solving them for turbulence needs very fine resolution or averaging.

  4. 4.Why are exact solutions to the Navier-Stokes equations important in fluid mechanics?Concept

    Exact solutions to the Navier-Stokes equations are important because they provide insights into the behavior of fluid flows under specific conditions. These solutions serve as benchmarks for testing numerical methods and approximations. They also help in understanding fundamental flow phenomena and validating experimental results.

  5. 5.Explain how the Navier-Stokes equations are applied in the design of aircraft.Application

    In aircraft design, the Navier-Stokes equations are used to predict the aerodynamic forces and moments acting on the aircraft. By solving these equations, engineers can simulate airflow over the aircraft's surfaces, optimizing the design for lift, drag, and stability. This helps in improving fuel efficiency, performance, and safety of the aircraft.

  6. 6.What happens if the viscosity term is neglected in the Navier-Stokes equations?Application

    Neglecting the viscosity term in the Navier-Stokes equations leads to the Euler equations, which describe inviscid flow. This simplification is valid for high Reynolds number flows where viscous effects are negligible compared to inertial forces. However, it fails to capture boundary layer effects, flow separation, and other phenomena where viscosity plays a crucial role.

  7. 7.Why is the Navier-Stokes equation considered challenging to solve analytically?Application

    The Navier-Stokes equations are challenging to solve analytically due to their nonlinearity and the complexity of boundary conditions in real-world problems. The equations involve coupled partial differential equations that can exhibit chaotic behavior, making it difficult to find general solutions. As a result, numerical methods and computational fluid dynamics (CFD) are often used to approximate solutions.

  8. 8.For a laminar flow between two parallel plates, derive the velocity profile using the Navier-Stokes equations.Numerical

    For laminar flow between two parallel plates (Couette flow), assume steady, incompressible flow with no pressure gradient. The Navier-Stokes equation simplifies to: d²u/dy² = 0. Integrating twice with respect to y gives: u(y) = Ay + B. Applying boundary conditions (u = U at y = 0 and u = 0 at y = h), we find A = -U/h and B = U. Thus, the velocity profile is u(y) = U(1 - y/h).

  9. 9.Discuss the role of boundary conditions in solving the Navier-Stokes equations.Application

    Boundary conditions are crucial in solving the Navier-Stokes equations as they define the behavior of the fluid at the boundaries of the domain. Common types include no-slip conditions at solid surfaces, where the fluid velocity matches the surface velocity, and free-slip conditions at interfaces. Properly defined boundary conditions ensure the uniqueness and physical relevance of the solution, influencing the flow pattern and stability.

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

Stuck on something here?