Pipe Flow — Reynolds Number & the Navier–Stokes Solution

Inertia versus viscosity

Two forces fight to set the personality of any fluid flow. Inertia wants the fluid to keep moving in whatever direction it's moving — it doesn't want to be redirected. Viscosity wants neighboring layers of fluid to stick together and damp out any differences. The ratio of those two is the most famous dimensionless number in fluid mechanics: the Reynolds number.

Source: Wikipedia — Reynolds number.

For pipe flow, the practical takeaway is a clean threshold:

• $\mathrm{Re} \lesssim 2300$ — viscosity wins. Flow is laminar: smooth parallel layers, perfectly predictable.
• $2300 \lesssim \mathrm{Re} \lesssim 4000$ — transitional. Tiny disturbances start to grow into eddies. Behavior gets unstable.
• $\mathrm{Re} \gtrsim 4000$ — inertia wins. Flow is turbulent: chaotic eddies, time-averaged but never steady at the small scale.

This is what the slider above is exposing. Try the Water through a thin tube preset (Re ≈ 500): the tracers march in clean parallel lanes. Now switch to the faucet preset (Re ≈ 25,000) — same fluid, same physics, but the tracers thrash. Nothing about the fluid changed; only the regime did.

Hagen–Poiseuille — an exact NS solution

The full Navier–Stokes equations are infamously hard:

Source: Wikipedia — Navier–Stokes equations. General closed-form solutions don't exist; that's literally a Millennium Prize Problem.

But for one specific case — steady, fully developed, axisymmetric flow in a long round pipe — every term you can drop, drops. The convective acceleration $(\mathbf{v}\!\cdot\!\nabla)\mathbf{v}$ vanishes, the time derivative vanishes, and what's left is a balance between the axial pressure gradient and viscous shear:

Integrate twice with the no-slip boundary condition ($u = 0$ at $r = R$) and you get a perfect parabolic profile:

The cross-section average is half the centerline peak, $u_{\max} = 2\bar{v}$, and integrating the profile across the pipe yields the Hagen–Poiseuille pressure drop:

Source: Wikipedia — Hagen–Poiseuille equation. Open the engineering view above to see the parabolic profile drawn explicitly.

Why doubling the diameter is so powerful

Look at the pressure-drop formula: $\Delta P \propto 1/D^{2}$ at fixed $\bar{v}$. But the flow rate relationship is even more dramatic. Because $Q = \bar{v} A = \bar{v} \pi R^{2}$, if you instead hold $\Delta P$ constant and ask how much fluid you can move:

The fourth power on $R$ is why your kitchen sink drain is so much more effective than a thin straw, and why narrowing of arteries (which shrinks $R$) so dramatically increases the heart's pumping work. Halve the radius and you cut the carrying capacity by a factor of 16.

Why the laminar formula breaks at high Re

Hagen–Poiseuille assumed the velocity profile is steady, axisymmetric, and parallel — every fluid particle moves in a straight line down the pipe. That's a perfectly good solution to NS, but at high Reynolds number it's an unstable one. Tiny disturbances (a wall bump, a vibration, the inlet) get amplified by the inertia of the flow instead of damped by viscosity. The neat parabola breaks up into eddies, the time-averaged profile flattens out into a "plug" with steep wall gradients, and pressure drop now scales roughly as $\bar{v}^{1.75}$ to $\bar{v}^{2}$ instead of linearly. None of the closed-form formulas above apply once you cross the transition.