Reynolds Number Atm⚠ unverified

Understanding the Parameters

• What $Re$ is — a ratio of two forces, not a property of the air. Large $Re$ means inertia dominates and the flow is prone to turbulence; small $Re$ means viscosity dominates and the flow is smooth, orderly, and reversible. It is dimensionless by construction, which is what makes it comparable across scales.

• Air density $\rho$ — supply the value at your flight altitude from ISA Density, not the sea-level 1.225. Climbing to the tropopause cuts $\rho$ to 30 % of sea level and cuts $Re$ in the same proportion at fixed speed.

• Flow speed $V$ — the freestream (true airspeed) relative to the body.

• Characteristic length $L$ — the choice that makes $Re$ meaningful, and the one most often botched. For a wing it is conventionally the chord (streamwise), not the span. For a pipe, the diameter; for a fuselage, the length. Comparing two Reynolds numbers computed on different reference lengths is meaningless, so the convention must always be stated. Doubling $L$ doubles $Re$.

• Dynamic viscosity $\mu$ — the air's resistance to shear, in $\text{Pa}\cdot\text{s}$. Note the trap: for gases $\mu$ rises with temperature (unlike liquids, where it falls), because gas viscosity comes from molecular momentum exchange, which speeds up when molecules do — roughly as $\mu \propto T^{3/2}/(T+110.4)$, Sutherland's law. The default $1.789\times10^{-5}$ is the ISA sea-level ($288.15\ \text{K}$) value; at the $216.65\ \text{K}$ tropopause it drops to about $1.42\times10^{-5}\ \text{Pa}\cdot\text{s}$, so the default is wrong at cruise — it is a starting point, not a constant. $\mu$ depends on temperature but is essentially independent of pressure, a fact that surprises most people and that kinetic theory explains (see the derivation).

• The combination $\nu = \mu/\rho$ — the kinematic viscosity (m²/s), letting $Re = VL/\nu$. Unlike $\mu$, $\nu$ is strongly pressure-dependent (through $\rho$), rising steeply with altitude: this is why high-altitude flight is a low-Reynolds regime.

• Typical values — a bacterium: $Re \sim 10^{-5}$; an insect wing: $10^2$–$10^4$; a model aircraft or a high-altitude UAV: $10^5$; a general-aviation wing: $10^6$; an airliner wing: $10^7$; a fuselage: $10^8$. Transition from laminar to turbulent on a flat plate occurs near $Re \approx 5\times10^5$.

• Why it dominates aerodynamics — Drag Force's $C_D$, the boundary-layer thickness, the separation point, and the stall behaviour are all functions of $Re$. This is why wind-tunnel testing is hard: a scale model at $1/10$ size has $1/10$ the $Re$ at the same speed and density, so it lies about the full-scale flow unless density or speed is raised to compensate — the entire reason pressurised and cryogenic wind tunnels exist.

Derivation (Approaching a Proof)

Route 1 — a ratio of forces. Estimate the two competing forces on a fluid element of size $L$ by scaling argument. Inertial force is mass × acceleration. Mass scales as $\rho L^3$; acceleration is the rate of velocity change a particle experiences traversing the body, $\sim V \cdot (V/L) = V^2/L$. So

$$F_{\text{inertia}} \sim \rho L^3 \cdot \frac{V^2}{L} = \rho V^2 L^2.$$

Viscous force is shear stress × area. Newton's law of viscosity gives stress $\tau = \mu\,dV/dy \sim \mu V/L$, acting over area $\sim L^2$:

$$F_{\text{viscous}} \sim \mu\frac{V}{L}\cdot L^2 = \mu V L.$$

The ratio is the Reynolds number:

$$Re = \frac{F_{\text{inertia}}}{F_{\text{viscous}}} \sim \frac{\rho V^2 L^2}{\mu V L} = \frac{\rho V L}{\mu}. \qquad\blacksquare$$

Route 2 — non-dimensionalising Navier–Stokes. The deeper justification. The incompressible momentum equation is

$$\rho\left(\frac{\partial \mathbf{u}}{\partial t} + \mathbf{u}\cdot\nabla\mathbf{u}\right) = -\nabla p + \mu\nabla^2\mathbf{u}.$$

Introduce dimensionless variables scaled on the problem's own units: $\mathbf{u}^\ast = \mathbf{u}/V$, $\mathbf{x}^\ast = \mathbf{x}/L$, $t^\ast = tV/L$, $p^\ast = p/(\rho V^2)$. Substituting and dividing through by $\rho V^2/L$ collapses every constant into a single group:

$$\frac{\partial \mathbf{u}^\ast}{\partial t^\ast} + \mathbf{u}^\ast\cdot\nabla^\ast\mathbf{u}^\ast = -\nabla^\ast p^\ast + \frac{\mu}{\rho V L}\nabla^{\ast 2}\mathbf{u}^\ast = -\nabla^\ast p^\ast + \frac{1}{Re}\nabla^{\ast 2}\mathbf{u}^\ast.$$

This is the profound statement, and the reason $Re$ matters more than any of its ingredients. The governing equation in dimensionless form contains $\rho$, $V$, $L$, $\mu$ only through the combination $Re$. Therefore any two flows with the same $Re$ and the same geometry obey the identical equation and have identical dimensionless solutions — this is dynamic similarity. A 1/10-scale model in a tunnel at 10× the speed reproduces the full-scale flow exactly. Without this theorem, wind tunnels would be decorative.

The formula is also recoverable by Buckingham π: with 4 variables ($\rho, V, L, \mu$) in 3 dimensions (M, L, T), $4 - 3 = 1$ dimensionless group exists — and it is $Re$. There is nothing else it could be.

Dimensional check. With $\text{Pa}\cdot\text{s} = \text{kg}/(\text{m}\cdot\text{s})$: $$\frac{\rho V L}{\mu} = \frac{(\text{kg}/\text{m}^3)(\text{m}/\text{s})(\text{m})}{\text{kg}/(\text{m}\cdot\text{s})} = \frac{\text{kg}/(\text{m}\cdot\text{s})}{\text{kg}/(\text{m}\cdot\text{s})} = 1\ \checkmark$$ Dimensionless, as a similarity parameter must be.

History and Development

• Viscosity formalised (1840s). Jean Léonard Marie Poiseuille's meticulous experiments on flow through fine capillaries (published 1840s, motivated by blood circulation) quantified viscous resistance; George Gabriel Stokes and Claude-Louis Navier built the governing equations that now bear their names.

• Reynolds's experiment (1883). Osborne Reynolds, at Owens College Manchester, injected a filament of dye into water flowing through a glass tube and varied the speed. At low speeds the dye drew a clean straight line down the tube; above a critical speed it burst into eddies and mixed. He showed the transition was governed not by speed alone but by the combination $\rho V D/\mu$ — the same number for water and other fluids, in tubes of different diameters. His 1883 paper is one of the most consequential in fluid mechanics, and his apparatus still exists in Manchester.

• Naming (1908). Reynolds himself did not name the group. Arnold Sommerfeld proposed calling it the Reynolds number in 1908, and the name stuck.

• Similarity and the boundary layer (1904– ). Ludwig Prandtl's 1904 boundary-layer concept explained why $Re$ decides everything: at high $Re$ viscosity is negligible except in a thin layer near the surface, and it is that layer's behaviour — attached or separated, laminar or turbulent — that sets drag and stall. Together, $Re$-similarity and boundary-layer theory turned aerodynamics from an empirical art into a science, and made model testing predictive.

• Still the binding constraint. Matching full-scale $Re$ in a tunnel remains so difficult that NASA built the National Transonic Facility, which chills nitrogen to cryogenic temperatures to raise $\rho$ and lower $\mu$ enough to reach flight Reynolds numbers on small models. Reynolds's dye filament still sets the budget.

Related Concepts: Reynolds Number, Reynolds Number Heat, ISA Density, ISA Temperature, Dynamic Pressure, Drag Force, Lift Force, Parasite Drag, Mach Number, Froude Number, Weber Number, Prandtl Number

Notes: Duplicate of Reynolds Number (Physics / Fluids) — identical formula; only difference is the default $\mu = 1.789\times10^{-5}\ \text{Pa}\cdot\text{s}$ (ISA sea-level air). Third instance: Reynolds Number Heat. Unit labels correct here. $\mu$ rises with temperature for gases (Sutherland's law) and is ~independent of pressure — the sea-level default is wrong at cruise ($\approx 1.42\times10^{-5}$ at the tropopause). $L$ convention (chord, not span) must always be stated. From non-dimensionalised Navier–Stokes: same $Re$ + same geometry ⇒ dynamically similar flows.