Flow Lab

Syllabus / Module 11: Fluid Dynamics: Navier–Stokes and Beyond

Lesson 11.4

Viscosity & the Navier–Stokes Equations

Stress tensor, Newtonian fluids, deriving the full Navier–Stokes equations, exact solutions (Couette, Poiseuille), the no-slip condition.

Written byClaude Opus

🌊
Why this matters for simulation: The equations your wind tunnel will solve.

By the end of this lesson you can…

  • Explain viscosity as momentum diffusing between layers and use $\tau = \mu\,du/dy$ with $\nu = \mu/\rho$ for air and water.
  • Build the stress tensor $\sigma = -pI + 2\mu S$ from a shear flow, and read the force on any face as $\sigma\mathbf n$.
  • Take the divergence of the stress to reach $\mu\nabla^2\mathbf u$ and write the Navier–Stokes equations with no-slip.
  • Solve Couette and Poiseuille flow as two-point BVPs and evaluate wall shear stress: 0.9 Pa under a plate sliding at 50 m/s over 1 mm of air.
  • Solve Stokes' first problem by similarity, $u = U\,\mathrm{erfc}\big(y/2\sqrt{\nu t}\big)$, and check it with finite differences stable only for $\nu\,\Delta t/\Delta y^2 \le 1/2$.
  • Compute the viscous dissipation $2\mu S\!:\!S$ and show it equals the work done by the walls.
On this page
·
  1. Viscosity: momentum leaking between layers
  2. Worked example: a plate over a thin film of air
  3. The stress tensor, built from a shear flow
  4. The Newtonian law
  5. From stress to Navier–Stokes
  6. No-slip
  7. Plate flows: Navier–Stokes as a two-point BVP
  8. Worked example: Poiseuille in a 1 cm slot of air
  9. Momentum diffuses like heat
  10. Stokes’ first problem: the impulsively started plate
  11. Checking erfc numerically, and the stability limit
  12. Energy dissipation
  13. Computing it
  14. Concept checks
  15. Problems
  16. Key Takeaways

The last lesson ended with a cylinder that feels no drag. Euler’s air slides along a surface as if oiled, forbidden only to pass through it. Real air sticks: at the skin of a wing it moves with the wing, and within a few millimetres its speed climbs to flight speed. The shearing in that thin layer is where skin-friction drag comes from, and where, much later, stall begins.

The missing physics is viscosity; adding it to Euler gives the Navier–Stokes equations, the equations a wind-tunnel simulator solves. One idea is new: momentum diffuses, exactly as heat does. The tools you own: the strain-rate tensor SS from 11.1 Continuum Hypothesis & Flow Kinematics, the momentum balance from 11.3 Momentum: Pressure & the Euler Equations, BVPs from 8.1 Two-Point Boundary Value Problems, and stability of explicit steps from 4.5 Stability Regions & Stiffness.

▶ Video Molecules carrying momentum between layers, the stress on a tilted face, plate flows from a BVP, an impulsively started plate and the erfc profile, and an explicit scheme that explodes past r = 1/2captions only
Transcript
  1. The last lesson ended with a cylinder that feels no drag: Euler's air slides along a surface as if oiled, which is d'Alembert's paradox. Real air sticks. At the skin of a wing it moves with the wing, and within a few millimetres its speed climbs to flight speed. That shearing is where skin friction comes from. The missing physics is viscosity.
  2. Picture two layers of air, the upper one faster. Molecules hop across the boundary carrying momentum: the slow layer is dragged forward, the fast one held back. Momentum per second per area is a shear stress, and Newton's law says tau equals mu du dy. Divide by density and you get nu, a diffusion coefficient: one point four seven times ten to the minus five for air, ten to the minus six for water. Water is fifty-five point six times stickier by mu, yet air's nu is fourteen point seven times larger.
  3. Slide a plate at fifty metres per second over a one millimetre film of air. The shear rate is five times ten to the fourth, so the stress is zero point nine pascals. To handle any face we need a tensor, sigma, that turns a face normal into the force on it. Shears on the top and bottom, balanced by the sides, plus pressure, give minus p on the diagonal and mu gamma off it: minus p I plus two mu S. Tilt the face to forty-five degrees and the force is purely normal: plus or minus zero point nine pascals.
  4. Take the divergence of that stress. Continuity kills the extra term and leaves mu times the Laplacian. Divide by rho and you have the Navier-Stokes equations: Euler plus nu del squared u. The new term is diffusion, so the equation is second order in space, and a wall needs a second condition. That is no-slip: the fluid's velocity equals the wall's.
  5. Plate flows turn Navier-Stokes into a two-point boundary value problem. With no pressure gradient, Couette is a straight line. Between fixed plates, Poiseuille is a parabola. For air in a one centimetre slot with G equal to one pascal per metre, the peak is zero point six nine four four metres per second, the mean zero point four six three zero, and the flow rate four point six three times ten to the minus three. Start it from rest: every layer first gains zero point eight one six, then the walls' brake diffuses inward, and the slowest mode decays at one point four five zero two per second.
  6. Now start air at rest and snap the plate to speed U. Momentum diffuses upward, and a similarity argument turns the heat equation into an ODE whose solution is U times erfc of y over two root nu t. The edge of the layer is four root nu t: four point eight five millimetres after a tenth of a second, fifteen point three after one second, growing like root t. The dots are a finite-difference run, and they sit on the curve.
  7. Why must r be small? Feed forward Euler the sharpest pattern a grid can hold, a sawtooth. Each step multiplies it by one minus four r. Up to r equal to one half that factor stays between minus one and one. At point four the error is eight point seven times ten to the minus four. At one half it is two point one times ten to the minus three and never decays. At point five one it has grown to zero point zero nine eight, and at point six, three times ten to the eleven.
  8. Where does the energy go? Only strain heats the fluid: the dissipation is two mu S double dot S. In the Couette film that is forty-five thousand watts per cubic metre, forty-five per square metre, exactly the plate's power: zero point nine pascals times fifty. Next lesson: the equations hold one viscosity, one speed and one length, but only one combination matters, the Reynolds number. For this film it is three thousand four hundred three.

Viscosity: momentum leaking between layers

Picture two layers of air, the upper faster. Molecules hop across the boundary at random (kinetic theory, from memory). One crossing downward carries the upper layer’s larger xx-momentum into the slow layer; one crossing upward brings a deficit. No net mass crosses, but momentum does: the slow layer is dragged forward, the fast one held back. Momentum per second per unit area is a force per unit area, a shear stress τ\tau.

For most everyday fluids it is proportional to the shear rate, Newton’s law of viscosity:

τ=μ dudy.\tau = \mu\,\frac{du}{dy} .

μ\mu is the dynamic viscosity (Pa s). Divided by density it gives the kinematic viscosity ν=μ/ρ\nu = \mu/\rho, in m2^2/s, the units of a diffusion coefficient. Sea-level values, approximate and from memory: air μ=1.8×10−5\mu = 1.8\times10^{-5} Pa s, so ν=1.8×10−5/1.225=1.47×10−5\nu = 1.8\times10^{-5}/1.225 = 1.47\times10^{-5} m2^2/s; water μ=1.0×10−3\mu = 1.0\times10^{-3}, ν=1.0×10−6\nu = 1.0\times10^{-6}. Water is 55.655.6 times stickier by μ\mu, yet air’s ν\nu is 14.714.7 times larger: air is so light that a little friction moves it a lot.

Worked example: a plate over a thin film of air

A plate slides at U=50U = 50 m/s over a 11 mm gap of air above a fixed floor, the air sheared linearly. Then du/dy=50/0.001=5×104du/dy = 50/0.001 = 5\times10^4 s−1^{-1} and

τ=1.8×10−5⋅5×104=0.9 Pa.\tau = 1.8\times10^{-5}\cdot5\times10^4 = 0.9 \text{ Pa}.

That is 5.9×10−45.9\times10^{-4} of the dynamic pressure, 12ρU2=1531.25\tfrac12\rho U^2 = 1531.25 Pa (11.3’s pitot reading). A tiny stress, but it acts along the flight direction over the whole wetted surface, while pressure forces largely cancel in that direction: hence skin friction is a major part of a clean airplane’s drag (from memory).

The stress tensor, built from a shear flow

In 11.3 a face felt only pressure, along −n-\mathbf n. Now it also feels a sideways pull. We want a machine taking a face’s normal n\mathbf n to the force per area on it, the traction t\mathbf t; it is a matrix, the stress tensor: t=σn\mathbf t = \sigma\mathbf n.

Build it for the shear flow u=(γy,0)\mathbf u = (\gamma y, 0) of 11.1, at pressure pp:

  • Top face, n=(0,1)\mathbf n = (0, 1): the faster air above drags it forward with μγ\mu\gamma; pressure pushes down with pp. So t=(μγ,−p)\mathbf t = (\mu\gamma, -p), the second column of σ\sigma.
  • Right face, n=(1,0)\mathbf n = (1, 0): pressure pushes left with pp. A shear too? On a tiny square parcel the top and bottom shears form a couple; it scales like the side squared while the moment of inertia scales like its fourth power, so the parcel would spin up infinitely fast unless the side faces carry an opposite couple. So the right face feels μγ\mu\gamma along yy: t=(−p,μγ)\mathbf t = (-p, \mu\gamma), the first column.
σ=(−pμγμγ−p)=−p I+2μ(0γ/2γ/20)=−p I+2μS.\sigma = \begin{pmatrix} -p & \mu\gamma \\ \mu\gamma & -p \end{pmatrix} = -p\,I + 2\mu\begin{pmatrix} 0 & \gamma/2 \\ \gamma/2 & 0 \end{pmatrix} = -p\,I + 2\mu S .

The second matrix is exactly this flow’s strain-rate tensor SS from 11.1. Now tilt the face to 45∘45^\circ, n=(1,1)/2\mathbf n = (1, 1)/\sqrt2, the eigenvector of SS along which the fluid stretches. Then σn=(−p+μγ) n\sigma\mathbf n = (-p + \mu\gamma)\,\mathbf n: a pure normal force, pressure reduced by μγ\mu\gamma, a viscous tension. On the 135∘135^\circ face, the squeezing direction, it is −p−μγ-p - \mu\gamma: extra compression. In the 11 mm film that is ±0.9\pm0.9 Pa. Shear and normal stress are one tensor seen from different faces.

The Newtonian law

For a general flow, a Newtonian incompressible fluid has

σ=−p I+2μS,S=12(L+LT),Lij=∂ui∂xj.\sigma = -p\,I + 2\mu S, \qquad S = \tfrac12\big(L + L^{\mathsf T}\big),\quad L_{ij} = \frac{\partial u_i}{\partial x_j}.

Stated, not derived, it assumes: viscous stress is linear in the velocity gradient (air and water, not paint or blood); it ignores the rotation WW, since rigid spinning rubs nothing; it is isotropic, no preferred direction, leaving one constant μ\mu; and it is symmetric, by the torque argument. Since tr⁡S=∇⋅u=0\operatorname{tr} S = \nabla\cdot\mathbf u = 0, pp is still minus the average normal stress. (A compressible gas adds a term in ∇⋅u\nabla\cdot\mathbf u; 11.12.)

From stress to Navier–Stokes

The box argument of 11.3 carries over: the xx-force on a box is the stress on its right face minus its left, plus top minus bottom, and by Taylor, per unit volume,

fx=∂σxx∂x+∂σxy∂y,f_x = \frac{\partial\sigma_{xx}}{\partial x} + \frac{\partial\sigma_{xy}}{\partial y} ,

the divergence of the first row of σ\sigma. With σxx=−p+2μux\sigma_{xx} = -p + 2\mu u_x and σxy=μ(uy+vx)\sigma_{xy} = \mu(u_y + v_x):

fx=−px+μ(2uxx+uyy+vxy)=−px+μ(uxx+uyy)+μ ∂∂x(ux+vy)⏟= 0.f_x = -p_x + \mu\big(2u_{xx} + u_{yy} + v_{xy}\big) = -p_x + \mu\big(u_{xx} + u_{yy}\big) + \mu\,\frac{\partial}{\partial x}\underbrace{\big(u_x + v_y\big)}_{=\,0} .

Continuity kills the last bracket and leaves the Laplacian (2.11 The Laplacian & Vector Identities). The yy-row works the same way. Newton’s law per unit volume, divided by ρ\rho, gives the incompressible Navier–Stokes equations:

∂u∂t+(u⋅∇)u=−1ρ∇p+ν ∇2u+g,∇⋅u=0.\frac{\partial\mathbf u}{\partial t} + (\mathbf u\cdot\nabla)\mathbf u = -\frac1\rho\nabla p + \nu\,\nabla^2\mathbf u + \mathbf g, \qquad \nabla\cdot\mathbf u = 0 .

One new term, ν∇2u\nu\nabla^2\mathbf u. The Laplacian measures how far a value sits below its neighbours’ average, so viscosity pulls each parcel’s velocity toward its neighbours’: diffusion. The count still closes and pressure is still the multiplier enforcing ∇⋅u=0\nabla\cdot\mathbf u = 0, but the equation is now second order in space, which changes the boundary conditions.

No-slip

At a solid wall the fluid velocity equals the wall’s, both components: u=uwall\mathbf u = \mathbf u_{\text{wall}}. Physically (from memory): molecules striking a microscopically rough wall bounce off in random directions, forgetting their tangential momentum, so the gas touching the wall shares its velocity on average. It fails only in rarefied gas, where the mean free path rivals the body size.

Euler, first order in space, takes one wall condition, u⋅n=0\mathbf u\cdot\mathbf n = 0. Navier–Stokes, second order, needs the tangential one too. As ν→0\nu \to 0 the highest derivative fades but the extra condition stays, so the solution must change abruptly in a thin layer to meet it. That layer is the boundary layer (11.8).

Plate flows: Navier–Stokes as a two-point BVP

Seek a flow u=(u(y),0)\mathbf u = (u(y), 0) between plates at y=0y = 0 (fixed) and y=hy = h (sliding at UU). Continuity holds automatically, and the nonlinear term u ux+v uyu\,u_x + v\,u_y vanishes identically: that is why these flows have exact solutions. The yy-equation gives py=0p_y = 0 (ignoring gravity), so pp depends on xx only, and the xx-equation is

dpdx=μ d2udy2.\frac{dp}{dx} = \mu\,\frac{d^2u}{dy^2} .

The left side depends only on xx, the right only on yy, so both equal a constant, −G-G with G=−dp/dxG = -dp/dx. The BVP μu′′=−G\mu u'' = -G, u(0)=0u(0) = 0, u(h)=Uu(h) = U, integrates twice:

u(y)=U yh+G2μ y (h−y).u(y) = U\,\frac{y}{h} + \frac{G}{2\mu}\,y\,(h - y) .

Couette (G=0G = 0): a straight line, the 11 mm film. Plane Poiseuille (U=0U = 0): a parabola, which central differences reproduce exactly at the nodes.

Worked example: Poiseuille in a 1 cm slot of air

G=1G = 1 Pa/m, h=1h = 1 cm. The peak, at mid-gap, is umax⁡=Gh2/(8μ)=10−4/(1.44×10−4)=0.6944u_{\max} = Gh^2/(8\mu) = 10^{-4}/(1.44\times10^{-4}) = 0.6944 m/s, the mean is two thirds of that, 0.46300.4630 m/s, and the flow rate per unit width is Gh3/(12μ)=4.63×10−3Gh^3/(12\mu) = 4.63\times10^{-3} m2^2/s. The wall shear is μu′(0)=Gh/2=0.005\mu u'(0) = Gh/2 = 0.005 Pa, for a reason: a slab of length ℓ\ell is pushed by GℓhG\ell h of pressure and held by 2τwℓ2\tau_w\ell of friction, and steady means they balance.

Momentum diffuses like heat

Start the channel from rest. With u=(u(y,t),0)\mathbf u = (u(y, t), 0) the nonlinear term still vanishes:

ut=Gρ+ν uyy.u_t = \frac{G}{\rho} + \nu\,u_{yy} .

Without the forcing this is the heat equation, Tt=κTyyT_t = \kappa T_{yy}, with velocity as temperature. Read it as a stencil, uyy≈(uj+1−2uj+uj−1)/Δy2u_{yy} \approx (u_{j+1} - 2u_j + u_{j-1})/\Delta y^2: a layer slower than its neighbours’ average speeds up, a faster one slows, at a rate set by ν\nu. Units fix the time scale: ν\nu is m2^2/s, so momentum crosses a gap hh in a time of order h2/νh^2/\nu. For 11 cm of air that is 10−4/1.47×10−5=6.8110^{-4}/1.47\times10^{-5} = 6.81 s, a scale, not a rate.

Separating variables gives the rate. Write u=us(y)−e(y,t)u = u_s(y) - e(y, t), usu_s steady; the error obeys et=νeyye_t = \nu e_{yy} with e=0e = 0 at both plates. Try e=sin⁡(nπy/h) e−λte = \sin(n\pi y/h)\,e^{-\lambda t}: −λ=−νn2π2/h2-\lambda = -\nu n^2\pi^2/h^2. Every mode decays, sharp ones fastest; the last is n=1n = 1, at rate π2ν/h2=1.4502\pi^2\nu/h^2 = 1.4502 s−1^{-1}, time constant 0.6900.690 s: the scale divided by π2\pi^2. Expanding the initial error, usu_s itself, in these sines gives the exact transient.

◆ Interactive Channel start-up: 40-cell finite differences against the exact transient and the steady Couette–Poiseuille profile, with the error on a log scale

Loading simulation…

At first every layer accelerates at G/ρ=0.816G/\rho = 0.816 m/s2^2; the brake at the walls diffuses inward. The centre passes half its final speed at about 0.50.5 s (the exercise’s Fourier series gives 0.34740.3474 m/s then). On the log plot the error turns straight: one mode survives. The readout measures its decay rate over the last second as 1.45131.4513 s−1^{-1}, against 1.45021.4502 for the PDE; the gap is mostly forward Euler’s factor 1−λΔt1 - \lambda\Delta t per step. The error drops below 1%1\% at 3.173.17 s.

Stokes’ first problem: the impulsively started plate

Air at rest fills y>0y \gt 0, and at t=0t = 0 the plate at y=0y = 0 jumps to speed UU.

ut=ν uyy,u(0,t)=U,u→0 as y→∞,u(y,0)=0.u_t = \nu\,u_{yy}, \qquad u(0, t) = U, \qquad u \to 0 \text{ as } y \to \infty, \qquad u(y, 0) = 0 .

Similarity. The problem contains no length except νt\sqrt{\nu t}, so the profile must be one fixed shape stretched by it: u=Uf(η)u = U f(\eta) with η=y/(2νt)\eta = y/(2\sqrt{\nu t}). By the chain rule, ηt=−η/(2t)\eta_t = -\eta/(2t) and ηy=1/(2νt)\eta_y = 1/(2\sqrt{\nu t}), so

ut=−Uf′ η2t,νuyy=Uf′′4t⟹f′′=−2ηf′.u_t = -U f'\,\frac{\eta}{2t}, \qquad \nu u_{yy} = \frac{U f''}{4t} \quad\Longrightarrow\quad f'' = -2\eta f' .

The PDE has become an ODE, separable in f′f': f′=Ce−η2f' = C e^{-\eta^2}. With f(0)=1f(0) = 1 and f(∞)=0f(\infty) = 0, using ∫0∞e−s2ds=π/2\int_0^\infty e^{-s^2}ds = \sqrt\pi/2,

u=U erfc(y2νt),erfc(η)=1−2π∫0ηe−s2 ds.u = U\,\mathrm{erfc}\Big(\frac{y}{2\sqrt{\nu t}}\Big), \qquad \mathrm{erfc}(\eta) = 1 - \frac{2}{\sqrt\pi}\int_0^\eta e^{-s^2}\,ds .

Thickness. At y=4νty = 4\sqrt{\nu t}, u/U=erfc(2)=0.0047u/U = \mathrm{erfc}(2) = 0.0047: call that the edge, δ\delta. In air δ\delta is 15.315.3 mm after 11 s, and it grows like t\sqrt t.

Wall shear. Differentiating, ∂u/∂y=−Ue−η2/πνt\partial u/\partial y = -U e^{-\eta^2}/\sqrt{\pi\nu t}, so τw=μU/πνt\tau_w = \mu U/\sqrt{\pi\nu t}: infinite at the jump, then decaying as the layer thickens.

Checking erfc numerically, and the stability limit

Discretise ut=νuyyu_t = \nu u_{yy} with forward Euler in time and central differences in space (FTCS):

ujn+1=ujn+r (uj+1n−2ujn+uj−1n),r=ν ΔtΔy2.u_j^{n+1} = u_j^n + r\,\big(u_{j+1}^n - 2u_j^n + u_{j-1}^n\big), \qquad r = \frac{\nu\,\Delta t}{\Delta y^2} .

Why must rr be small? Feed it the sharpest pattern a grid can hold, the sawtooth uj=(−1)ju_j = (-1)^j. Then uj+1−2uj+uj−1=−4uju_{j+1} - 2u_j + u_{j-1} = -4u_j, and each step multiplies the sawtooth by 1−4r1 - 4r. For r≤1/2r \le 1/2 that factor lies in [−1,1][-1, 1]; above, the sawtooth flips sign and grows every step. The stiffness lesson said the same: the discrete Laplacian has eigenvalues down to −4ν/Δy2-4\nu/\Delta y^2, and forward Euler needs Δt ∣λ∣≤2\Delta t\,|\lambda| \le 2. Halve Δy\Delta y and the step must shrink fourfold.

◆ Interactive Stokes' first problem in air: FTCS dots against the erfc curve, with the stability ratio r as a slider

Loading simulation…

With Δy=0.5\Delta y = 0.5 mm, run to t≈1t \approx 1 s. At r=0.40r = 0.40 (Δt=6.81\Delta t = 6.81 ms, 147147 steps) the largest error is 8.7×10−4 U8.7\times10^{-4}\,U. At r=0.50r = 0.50 it is 2.1×10−32.1\times10^{-3}: the jump at the wall seeded a sawtooth with factor exactly −1-1, which never decays. At r=0.51r = 0.51 the factor is −1.04-1.04 and the error has grown to 0.0980.098; at r=0.60r = 0.60, factor −1.40-1.40, it reaches 3×10113\times10^{11} in 9898 steps.

Energy dissipation

The power of the viscous stress τij=2μSij\tau_{ij} = 2\mu S_{ij} per unit volume, ∂j(uiτij)\partial_j(u_i\tau_{ij}), splits in two: ui ∂jτiju_i\,\partial_j\tau_{ij} changes the kinetic energy, and the rest, τij ∂jui=2μSij(Sij+Wij)\tau_{ij}\,\partial_j u_i = 2\mu S_{ij}(S_{ij} + W_{ij}), heats the fluid. Since SS is symmetric and WW antisymmetric, SijWijS_{ij}W_{ij} sums to zero (pairs cancel), so the heating rate per unit volume is

Φ=2μ S ⁣: ⁣S=2μ∑i,jSij2 ≥0.\Phi = 2\mu\,S\!:\!S = 2\mu\sum_{i,j} S_{ij}^2 \ \ge 0 .

Only strain dissipates. For Couette, Sxy=Syx=γ/2S_{xy} = S_{yx} = \gamma/2, so S ⁣: ⁣S=γ2/2S\!:\!S = \gamma^2/2 and Φ=μγ2\Phi = \mu\gamma^2. In the 11 mm film, 1.8×10−5⋅(5×104)2=450001.8\times10^{-5}\cdot(5\times10^4)^2 = 45000 W/m3^3: 4545 W per square metre of plate. Check: the plate pushes against τ=0.9\tau = 0.9 Pa at 5050 m/s, 0.9⋅50=450.9\cdot50 = 45 W/m2^2. If the heat stayed, the film would warm at 36.636.6 K/s (cp=1005c_p = 1005 J/(kg K), from memory).

Computing it

🐍 PythonLab: viscosities (A), Couette (B), Poiseuille (C), Stokes' first problem (D), FTCS stability (E)
⌘/Ctrl + Enter
🧪 Coding exerciseExercise: start the channel from rest with explicit finite differences and watch it reach Couette–Poiseuille
⌘/Ctrl + Enter

The solution prints 0.34770.3477 m/s at t=0.5t = 0.5 s against the series’ 0.34740.3474 (graded to 0.0020.002 m/s) and 0.69440.6944 at 88 s, when the slowest mode is down to about 10−510^{-5} (graded to 0.1%0.1\% of umax⁡u_{\max}).

Concept checks

✅ Concept check

Air ($\mu = 1.8\times10^{-5}$ Pa s) flows steadily between fixed plates 1 cm apart under $G = 1$ Pa/m. What is the wall shear stress in Pa?

✅ Concept check

Why can Euler flow slip along a wall while Navier–Stokes flow cannot?

✅ Concept check

FTCS for diffusion in air ($\nu = 1.4694\times10^{-5}$ m$^2$/s) with $\Delta y = 0.5$ mm. What is the largest stable time step, in milliseconds?

✅ Concept check

An impulsively started plate drags a layer of thickness $\delta$ after time $t$. How long until the layer is twice as thick?

✅ Concept check

Which part of the motion makes viscous heat?

Problems

✏️ Pen & paperWhen does the wall shear vanish?●○○

Plates 11 cm apart, air, top plate sliding at U=0.5U = 0.5 m/s. (a) Find the wall shear τw=μu′(0)\tau_w = \mu u'(0) for the Couette–Poiseuille profile. (b) For what GG is it zero? (c) What happens below that?

Show worked solution

(a) u′(0)=U/h+Gh/(2μ)u'(0) = U/h + Gh/(2\mu), so τw=μU/h+Gh/2\tau_w = \mu U/h + Gh/2.

(b) G=−2μU/h2=−2⋅1.8×10−5⋅0.5/10−4=−0.18G = -2\mu U/h^2 = -2\cdot1.8\times10^{-5}\cdot0.5/10^{-4} = -0.18 Pa/m.

(c) u′(0)<0u'(0) \lt 0: backflow beside the fixed plate. With U=0.50U = 0.50 the channel panel’s steady wall shear is +4.0×10−4+4.0\times10^{-4} Pa at G=−0.1G = -0.1 and −1.0×10−4-1.0\times10^{-4} Pa at G=−0.2G = -0.2.

✏️ Pen & paperThe pump pays for the heat●●○

For plane Poiseuille flow, show that ∫0hμ u′(y)2 dy=GQ\int_0^h \mu\,u'(y)^2\,dy = GQ, where Q=∫0hu dyQ = \int_0^h u\,dy. Evaluate for G=1G = 1 Pa/m, h=1h = 1 cm, air.

Show worked solution

u=G2μy(h−y)u = \tfrac{G}{2\mu}y(h - y), u′=G2μ(h−2y)u' = \tfrac{G}{2\mu}(h - 2y). Then ∫0h(h−2y)2dy=h3/3\int_0^h (h - 2y)^2dy = h^3/3, so ∫μu′2dy=G2h312μ\int\mu u'^2dy = \tfrac{G^2h^3}{12\mu}. And Q=G2μ(h32−h33)=Gh312μQ = \tfrac{G}{2\mu}\big(\tfrac{h^3}{2} - \tfrac{h^3}{3}\big) = \tfrac{Gh^3}{12\mu}, so the dissipation equals GQGQ, the pressure drop’s pumping power. Numerically 4.63×10−34.63\times10^{-3} W per square metre of plate (lab block C).

✏️ Pen & paperStokes' first problem by hand●●●

(a) Verify that u=U erfc(y/2νt)u = U\,\mathrm{erfc}(y/2\sqrt{\nu t}) satisfies ut=νuyyu_t = \nu u_{yy} and the boundary conditions. (b) Find the wall shear stress. (c) For air and a plate at 5050 m/s, evaluate δ\delta and τw\tau_w at t=0.1t = 0.1 s.

Show worked solution

(a) For f=erfcf = \mathrm{erfc}, f′=−2πe−η2f' = -\tfrac{2}{\sqrt\pi}e^{-\eta^2}, so f′′=−2ηf′f'' = -2\eta f', which is what ut=νuyyu_t = \nu u_{yy} reduced to. At y=0y = 0, erfc(0)=1\mathrm{erfc}(0) = 1; as y→∞y \to \infty or t→0+t \to 0^+ with y>0y \gt 0, η→∞\eta \to \infty and erfc→0\mathrm{erfc} \to 0.

(b) uy=Uf′(η)/(2νt)u_y = U f'(\eta)/(2\sqrt{\nu t}), which at y=0y = 0 is −U/πνt-U/\sqrt{\pi\nu t}: τw=μU/πνt\tau_w = \mu U/\sqrt{\pi\nu t}, holding the plate back.

(c) νt=1.4694×10−6=1.212×10−3\sqrt{\nu t} = \sqrt{1.4694\times10^{-6}} = 1.212\times10^{-3} m, so δ=4.85\delta = 4.85 mm and τw=9×10−4/π⋅1.4694×10−6=0.419\tau_w = 9\times10^{-4}/\sqrt{\pi\cdot1.4694\times10^{-6}} = 0.419 Pa (lab block D).

Key Takeaways

  • Viscosity is momentum diffusion: τ=μ du/dy\tau = \mu\,du/dy; ν=μ/ρ\nu = \mu/\rho is 1.47×10−51.47\times10^{-5} m2^2/s for air, 14.714.7 times water’s.
  • σ=−pI+2μS\sigma = -pI + 2\mu S turns a face normal into a force: shear μγ\mu\gamma on flat faces, normal stresses ±μγ\pm\mu\gamma at 45∘45^\circ.
  • Its divergence is −∇p+μ∇2u-\nabla p + \mu\nabla^2\mathbf u: Navier–Stokes is Euler plus ν∇2u\nu\nabla^2\mathbf u, second order, so walls need no-slip.
  • Plate flows are BVPs: Couette Uy/hUy/h (0.90.9 Pa at 5050 m/s over 11 mm of air), Poiseuille G2μy(h−y)\tfrac{G}{2\mu}y(h - y).
  • Start-up is the heat equation; the slowest mode decays at π2ν/h2=1.45\pi^2\nu/h^2 = 1.45 s−1^{-1}, 1%1\% at 3.173.17 s.
  • Impulsive plate: u=U erfc(y/2νt)u = U\,\mathrm{erfc}(y/2\sqrt{\nu t}), δ=4νt\delta = 4\sqrt{\nu t}; FTCS reproduces it only for r≤1/2r \le 1/2.
  • Dissipation is 2μS ⁣: ⁣S2\mu S\!:\!S: 4545 W/m2^2 in the Couette film, exactly the plate’s power τU\tau U.
  • Next, 11.5: the equations hold a viscosity ν\nu, a speed UU and a length LL, yet only one combination matters, the Reynolds number UL/νUL/\nu. For the 11 mm film it is 34033403, and whether that film stays smooth depends on it alone.
Finished this lesson?

Mark it complete to track your progress (saved in this browser). Reading every section does this for you.