Dimensionless Numbers in Heat Transfer

Why This Matters

Dimensionless numbers are the universal language of heat transfer analysis. They let engineers compare systems of vastly different scales and predict thermal behavior without solving the full governing equations every time. The core skill here is recognizing what physical phenomena each number captures and when to apply each one. Whether you're analyzing a heat exchanger, predicting flow transition, or determining if a simple lumped analysis will work, these numbers are your diagnostic tools.

Don't fall into the trap of memorizing definitions in isolation. The real skill is understanding which forces or resistances each number compares, how numbers relate to each other (like how Rayleigh combines Grashof and Prandtl), and what the magnitude tells you about system behavior. Master the physical intuition behind each ratio, and you'll handle any problem that comes up.

---

Flow Regime and Momentum Transport

These numbers characterize how fluid moves and whether inertial or viscous effects dominate. They form the foundation for all convective heat transfer analysis.

Reynolds Number (Re)

• Ratio of inertial to viscous forces: $Re = \frac{\rho V L}{\mu} = \frac{V L}{\nu}$ where $L$ is the characteristic length (pipe diameter, plate length, etc.)
• Flow regime indicator: for internal pipe flow, $Re < 2300$ is laminar, $2300 < Re < 4000$ is transitional, and $Re > 4000$ is turbulent. Note that these thresholds differ for other geometries (e.g., flow over a flat plate transitions around $Re \approx 5 \times 10^5$).
• Central to convection correlations: nearly every forced convection Nusselt correlation includes $Re$ as a primary variable.

Prandtl Number (Pr)

• Ratio of momentum diffusivity to thermal diffusivity: $Pr = \frac{\nu}{\alpha} = \frac{\mu c_p}{k}$. This is a fluid property only, independent of flow conditions or geometry.
• Boundary layer relationship: determines whether the velocity or thermal boundary layer is thicker. When $Pr > 1$, the thermal boundary layer is thinner than the velocity boundary layer (momentum diffuses faster than heat). When $Pr < 1$, the reverse is true.
• Typical values: liquid metals have $Pr \ll 1$ (e.g., mercury $Pr \approx 0.025$), gases typically $Pr \approx 0.7$, water at room temperature $Pr \approx 7$, and oils have $Pr \gg 1$ (often 100–10,000+).

---

Convective Heat Transfer Performance

These numbers quantify how effectively heat moves from surfaces to fluids. They're the outputs you're solving for in most convection problems.

Nusselt Number (Nu)

• Ratio of convective to conductive heat transfer: $Nu = \frac{hL}{k_f}$, where $k_f$ is the fluid thermal conductivity. It's essentially a dimensionless heat transfer coefficient.
• Higher values mean stronger convection. Turbulent flow, enhanced surfaces, and favorable geometry all increase $Nu$. A value of $Nu = 1$ would mean pure conduction through a stagnant fluid layer.
• The target variable in most convection problems. Correlations express $Nu = f(Re, Pr)$ for forced convection or $Nu = f(Ra)$ for natural convection.

Stanton Number (St)

• Ratio of heat transferred to the fluid to its thermal capacity: $St = \frac{h}{\rho V c_p} = \frac{Nu}{Re \cdot Pr}$
• Useful in heat exchanger and external flow analysis because it directly relates wall heat flux to bulk fluid energy transport.
• Connects momentum and heat transfer through the Reynolds analogy: $St \approx \frac{C_f}{2}$ when $Pr \approx 1$, where $C_f$ is the skin friction coefficient. This is powerful because it lets you estimate heat transfer from friction data.

Peclet Number (Pe)

• Ratio of advective to diffusive heat transport: $Pe = Re \cdot Pr = \frac{VL}{\alpha}$
• Large $Pe$ means convection dominates and thermal diffusion is negligible compared to bulk fluid motion.
• Practical simplification: when $Pe \gg 1$, axial conduction terms can often be dropped from the energy equation, which significantly simplifies analysis.

---

Natural Convection and Buoyancy Effects

When temperature differences drive fluid motion through density variations, buoyancy forces replace imposed velocity as the driving mechanism. There's no fan or pump here; gravity and thermal expansion do the work.

Grashof Number (Gr)

• Ratio of buoyancy to viscous forces: $Gr = \frac{g \beta \Delta T L^3}{\nu^2}$, where $\beta$ is the volumetric thermal expansion coefficient and $\Delta T$ is the temperature difference between the surface and the bulk fluid.
• The natural convection analog to $Re^2$: larger temperature differences and longer surfaces increase $Gr$, strengthening buoyancy-driven flow.
• Flow stability indicator: critical $Gr$ values determine transition from laminar to turbulent natural convection (for a vertical plate, transition occurs around $Gr \approx 10^9$).

Rayleigh Number (Ra)

• Product of Grashof and Prandtl: $Ra = Gr \cdot Pr = \frac{g \beta \Delta T L^3}{\nu \alpha}$
• Convection onset criterion: for a horizontal fluid layer heated from below, convective cells (Bénard cells) form when $Ra > 1708$. Below this threshold, heat transfers by conduction alone.
• Primary correlation variable for natural convection. Most $Nu$ correlations use $Ra$ directly rather than $Gr$ alone, because $Ra$ captures both the buoyancy driving force and the thermal diffusion resistance.

---

Transient Conduction Analysis

These numbers govern time-dependent heat transfer in solids. They're essential for heating/cooling process design and for determining which solution method to use.

Biot Number (Bi)

• Ratio of internal conduction resistance to external convection resistance: $Bi = \frac{hL_c}{k_s}$, where $L_c$ is the characteristic length of the solid (typically volume/surface area) and $k_s$ is the solid's thermal conductivity.
• Lumped capacitance criterion: if $Bi < 0.1$, internal temperature gradients are negligible. The entire solid is approximately at one uniform temperature, and a simple exponential decay solution applies.
• Physical meaning: small $Bi$ means the solid conducts heat internally much faster than the surface can transfer it to the surrounding fluid. The bottleneck is at the surface, not inside the solid.

Be careful not to confuse $Bi$ with $Nu$. They look similar ($\frac{hL}{k}$), but $Bi$ uses the solid's conductivity $k_s$ and characterizes resistance within the solid, while $Nu$ uses the fluid's conductivity $k_f$ and characterizes convective enhancement.

Fourier Number (Fo)

• Dimensionless time for conduction: $Fo = \frac{\alpha t}{L^2}$, where $\alpha$ is the thermal diffusivity of the solid. It represents the ratio of the rate of heat conduction to the rate of thermal energy storage.
• Measures thermal penetration: larger $Fo$ means heat has diffused further into the solid relative to its size. A small $Fo$ means you're early in the transient and the interior hasn't "felt" the surface condition yet.
• Appears in all transient solutions: Heisler charts, one-term approximation series, semi-infinite solid solutions, and numerical schemes all use $Fo$.

---

Special Applications

These numbers address specific phenomena that become important under particular conditions, such as high-speed flows and significant viscous heating.

Eckert Number (Ec)

• Ratio of flow kinetic energy to the enthalpy difference: $Ec = \frac{V^2}{c_p \Delta T}$
• Viscous dissipation indicator: when $Ec$ is significant, frictional heating within the fluid affects temperature profiles and can't be ignored in the energy equation.
• Critical for high-speed flows: compressible aerodynamics, high-velocity lubricant films, and polymer extrusion are common cases. For most low-speed engineering flows, $Ec \ll 1$ and viscous dissipation is safely neglected.

---

Quick Reference Table

---

Self-Check Questions

1. You're analyzing natural convection from a vertical plate. Which two dimensionless numbers would appear in your Nusselt correlation, and how are they related to each other?

2. A small steel sphere is quenched in oil. What dimensionless number determines whether you can use lumped capacitance analysis, and what threshold value must it satisfy?

3. Compare and contrast Reynolds number and Grashof number: what role does each play in forced vs. natural convection, and what physical forces does each ratio represent?

4. If a problem asks you to evaluate heat exchanger performance and compare different flow velocities, which dimensionless number directly relates heat transfer to the fluid's thermal capacity rate?

5. Two fluids have the same Reynolds number in identical tubes, but one is liquid metal ($Pr = 0.01$) and one is oil ($Pr = 1000$). How would their thermal boundary layers differ, and which would have the higher Nusselt number for the same $Re$?