Dimensionless Numbers in Heat Transfer: Nusselt, Reynolds, Prandtl, Grashof, and Rayleigh

Industry Applications

HVAC heat exchangers, nuclear reactor cooling, semiconductor thermal management, aerospace thermal protection, solar thermal collectors

Key Standards

ASHRAE Fundamentals Handbook (Ch. 22), VDI Heat Atlas (2nd ed.), Incropera & DeWitt Fundamentals of Heat and Mass Transfer

Typical Scale

Microelectronics: Re ~ 1–100, Nu ~ 5–50; Power plant condensers: Re ~ 10⁵–10⁶, Nu ~ 1000–5000

⚠️ Why It Matters

1

Incorrect Nusselt number estimation

↓

2

Underpredicted convective heat transfer coefficient

↓

3

Insufficient cooling capacity

↓

4

Thermal runaway in electronics or reactors

↓

5

Catastrophic equipment failure or safety violation

📘 Definition

Dimensionless numbers in heat transfer are non-dimensional groupings of physical properties and flow variables that characterize the relative importance of key transport mechanisms (e.g., convection vs. conduction, momentum vs. thermal diffusion). They form the basis for similarity analysis, empirical correlation development, and scaling of thermal systems across geometry, fluid, and operating conditions.

🎨 Concept Diagram

AI-generated illustration for visual understanding

💡 Engineering Insight

Never treat Nu as an output alone — it’s the *bridge* between fluid mechanics and thermodynamics. A mismatch between predicted and measured Nu almost always traces to incorrect film temperature selection or unaccounted property variation (e.g., using bulk instead of arithmetic mean temperature for Pr). Always compute properties at T_film = (T_s + T_∞)/2 for external flow, or T_film = (T_in + T_out)/2 for internal flow with moderate ΔT.

📖 Detailed Explanation

Dimensionless numbers arise from dimensional analysis (Buckingham Pi theorem), reducing complex partial differential equations into tractable functional relationships. For example, the energy equation simplifies to Nu = f(Re, Pr) for forced convection — meaning all geometrically similar flows with identical Re and Pr will have identical Nu, regardless of scale or fluid.

Beyond similarity, these numbers encode physics: Re governs turbulence onset and mixing; Pr reveals whether heat diffuses faster than momentum (low Pr, e.g., liquid metals) or vice versa (high Pr, e.g., oils); Gr and Ra quantify buoyancy-driven instability thresholds. Their combinations define correlation domains — e.g., the Sieder–Tate correction accounts for variable viscosity effects when μ_s/μ_b deviates significantly from unity.

At advanced levels, modern applications demand hybrid treatment: machine learning surrogate models trained on high-fidelity DNS data now predict Nu across wide Re–Pr–Ra spaces, but they still rely on dimensionless inputs for generalizability. Also, non-Newtonian fluids, nanofluids, and rotating systems introduce additional groups (e.g., Weissenberg, Hartmann, Ekman numbers), requiring extension of classical frameworks while preserving dimensional rigor.

🔄 Engineering Workflow

Step 1

Step 1: Identify dominant heat transfer mode (forced/natural convection, conduction, radiation)

→

Step 2

Step 2: Select governing dimensionless groups based on geometry, driving force, and fluid properties

→

Step 3

Step 3: Compute Re, Pr, Gr (or Ra), and Nu using measured or tabulated fluid properties at film temperature

→

Step 4

Step 4: Validate regime applicability (e.g., Re range for correlation, Ra threshold for convection onset)

→

Step 5

Step 5: Apply appropriate empirical or analytical correlation to estimate h or q''

→

Step 6

Step 6: Iterate with thermal resistance network or CFD boundary condition assignment

→

Step 7

Step 7: Verify against experimental data or benchmark simulations (e.g., ASHRAE RP-1145 validation cases)

📋 Decision Guide

Rock/Field Condition	Recommended Design Action
High Re (>10⁴) + Moderate Pr (~0.7–7) + Forced flow	Use Dittus–Boelter correlation (Nu = 0.023 Re⁰·⁸ Prⁿ, n=0.4 heating/0.3 cooling)
Low Re (<2300) + High Pr (>100) + Laminar internal flow	Apply Graetz solution or constant-surface-temperature correlations with entrance-length correction
Ra > 10⁹ in vertical enclosure (e.g., solar chimney, electronics cabinet)	Adopt Churchill–Chu correlation for vertical plates (Nu = 0.68 + 0.67 Ra^(1/4) / [1 + (0.492/Pr)^(9/16)]^(4/9))
Pr < 0.01 (liquid metals) + Turbulent flow	Use Seban–McLaughlin correlation (Nu = 5.0 + 0.025 Re⁰·⁸ Pr⁰·⁸)

📊 Key Properties & Parameters

Nusselt Number (Nu)

0.5–10,000 (laminar pipe flow: 3.6–60; turbulent pipe flow: 100–10,000; natural convection on vertical plates: 10–1000)

Ratio of convective to conductive heat transfer across a boundary; quantifies enhancement of heat transfer due to fluid motion.

⚡ Engineering Impact:

Directly determines required heat exchanger surface area and pumping power.

Reynolds Number (Re)

Laminar: < 2300 (pipes), Transitional: 2300–4000, Turbulent: > 4000 (pipes); External flow: > 5×10⁵ (flat plate transition)

Ratio of inertial to viscous forces; indicates whether flow is laminar, transitional, or turbulent.

⚡ Engineering Impact:

Dictates pressure drop, mixing efficiency, and applicability of correlation families (e.g., Dittus–Boelter vs. Sieder–Tate).

Prandtl Number (Pr)

Liquid metals: 0.004–0.03; Water: 1–13 (20°C→100°C); Air: 0.7–0.72; Oils: 50–100,000

Ratio of momentum diffusivity (kinematic viscosity) to thermal diffusivity; measures relative thickness of velocity and thermal boundary layers.

⚡ Engineering Impact:

Controls whether thermal or hydrodynamic boundary layer dominates — critical for selecting appropriate Nu–Re–Pr correlations.

Grashof Number (Gr)

Free convection over vertical plates: 10⁴–10¹²; Enclosures: 10³–10⁸

Ratio of buoyancy to viscous forces; quantifies strength of natural convection currents driven by density gradients.

⚡ Engineering Impact:

Determines onset and intensity of natural convection — essential for passive cooling design and furnace draft analysis.

Rayleigh Number (Ra)

Onset of convection in fluids: Ra > 1708 (horizontal layer); Vertical plates: 10⁴–10¹²; Enclosed cavities: 10³–10⁷

Product of Grashof and Prandtl numbers; represents combined effect of buoyancy and thermal diffusivity in natural convection.

⚡ Engineering Impact:

Defines stability of stratified fluids and transition between conduction-dominated and convection-dominated regimes.

📐 Key Formulas

Nusselt Number

Nu = h L_c / k

Defines convective heat transfer coefficient h relative to conductive resistance k/L_c

Variables:

Symbol	Name	Unit	Description
Nu	Nusselt Number	dimensionless	Dimensionless number representing the ratio of convective to conductive heat transfer
h	Convective Heat Transfer Coefficient	W/(m²·K)	Coefficient quantifying the rate of heat transfer between a solid surface and a fluid
L_c	Characteristic Length	m	Representative physical length scale of the system, e.g., diameter for a cylinder or length for a flat plate
k	Thermal Conductivity	W/(m·K)	Material property measuring ability to conduct heat

Typical Ranges:

Laminar pipe flow (fully developed)

3.6–60

Turbulent pipe flow (Dittus–Boelter)

100–10,000

Natural convection on vertical plate

10–1000

⚠️ Nu < 1 implies conduction dominates; Nu > 10⁴ suggests possible measurement error or unmodeled turbulence

Reynolds Number

Re = ρ V L_c / μ

Quantifies flow regime based on inertia vs. viscosity

Variables:

Symbol	Name	Unit	Description
ρ	Fluid density	kg/m³	Mass per unit volume of the fluid
V	Characteristic velocity	m/s	Typical flow velocity, e.g., freestream or mean velocity
L_c	Characteristic length	m	Representative physical length scale of the system, e.g., pipe diameter or chord length
μ	Dynamic viscosity	Pa·s	Measure of a fluid's resistance to shear deformation

Typical Ranges:

Laminar flow in circular pipes

0–2300

Turbulent flow in circular pipes

4000–10⁷

External flow over turbine blades

10⁵–10⁷

⚠️ Re < 1 indicates creeping flow (Stokes regime); Re > 10⁸ may require compressibility corrections

Prandtl Number

Pr = ν / α = c_p μ / k

Compares momentum and thermal diffusion rates

Variables:

Symbol	Name	Unit	Description
Pr	Prandtl Number	dimensionless	Dimensionless number comparing momentum diffusivity (kinematic viscosity) to thermal diffusivity
ν	kinematic viscosity	m²/s	Momentum diffusivity
α	thermal diffusivity	m²/s	Thermal diffusivity
c_p	specific heat capacity at constant pressure	J/(kg·K)	Heat capacity per unit mass at constant pressure
μ	dynamic viscosity	Pa·s	Absolute viscosity
k	thermal conductivity	W/(m·K)	Ability of a material to conduct heat

Typical Ranges:

Air at 25°C

0.71

Water at 20°C

7.0

Engine oil at 40°C

2000–10,000

⚠️ Pr < 0.01 requires liquid metal-specific correlations; Pr > 10⁵ invalidates standard boundary layer assumptions

Grashof Number

Gr = g β (T_s − T_∞) L_c³ / ν²

Measures buoyancy-driven flow strength in natural convection

Variables:

Symbol	Name	Unit	Description
Gr	Grashof Number	dimensionless	Dimensionless number representing the ratio of buoyancy to viscous forces in natural convection
g	Acceleration due to gravity	m/s²	Gravitational acceleration
β	Thermal expansion coefficient	1/K	Volumetric thermal expansion coefficient of the fluid
T_s	Surface temperature	K	Temperature of the solid surface
T_∞	Ambient fluid temperature	K	Temperature of the surrounding fluid far from the surface
L_c	Characteristic length	m	Representative physical length scale of the system
ν	Kinematic viscosity	m²/s	Ratio of dynamic viscosity to fluid density

Typical Ranges:

Vertical plate, air, ΔT = 20 K

10⁷–10⁸

Enclosed cavity, water, ΔT = 5 K

10⁵–10⁷

⚠️ Gr < 10³ implies negligible natural convection; Gr > 10¹² may trigger turbulent plume instabilities

Rayleigh Number

Ra = Gr · Pr

Combined buoyancy–diffusivity parameter for natural convection onset and regime

Variables:

Symbol	Name	Unit	Description
Ra	Rayleigh Number	dimensionless	Combined buoyancy–diffusivity parameter for natural convection onset and regime
Gr	Grashof Number	dimensionless	Ratio of buoyancy to viscous forces
Pr	Prandtl Number	dimensionless	Ratio of momentum diffusivity to thermal diffusivity

Typical Ranges:

Horizontal fluid layer (onset of convection)

1708

Vertical plate, air, ΔT = 30 K

10⁹–10¹⁰

⚠️ Ra < 10³ → pure conduction; Ra > 10¹³ → fully turbulent natural convection (rare in engineering practice)

🏭 Engineering Example

AP1000 Passive Containment Cooling System (Vogtle Unit 3, Georgia, USA)

N/A — Fluid system: Pressurized water (primary coolant) + atmospheric air (external cooling)

Gr

1.8 × 10⁹ (air-side natural convection, 40°C ΔT)

Nu

840 (measured via thermocouple arrays on containment steel shell)

Pr

1.32 (water at 300°C)

Ra

2.4 × 10⁹ (air, vertical containment wall, 30 m height)

Re

2.1 × 10⁵ (internal pipe flow, 300°C water)

🏗️ Applications

Thermal design of shell-and-tube heat exchangers
Cooling system validation for data center racks
Passive decay heat removal in Gen III+ nuclear reactors
Aerothermal modeling of hypersonic vehicle leading edges

📋 Real Project Case

Air-Cooled Condenser Retrofit for 600 MW Coal Power Plant

Retrofit of legacy water-cooled condenser at Midwest US plant

Challenge: Water scarcity forcing shift to dry cooling; risk of summer turbine backpressure rise

Read full case study →

Thermal Management System for EV Traction Inverter Furnace Refractory Lining Failure Analysis in Aluminum Melting Facility HVAC Coil Frost Detection and Defrost Optimization for Cold Storage Warehouse Thermal Design of Satellite Payload Radiator for Lunar Orbit Mission

🎨 Technical Diagrams

📚 References

[1]

ASHRAE Fundamentals Handbook — American Society of Heating, Refrigerating and Air-Conditioning Engineers

[2]

VDI Heat Atlas — Verein Deutscher Ingenieure

[3]

Heat Transfer — McGraw-Hill Education

[4]

Nuclear Systems Volume I: Thermal Hydraulic Fundamentals — American Nuclear Society

🔗 Related Knowledge

Newton’s Law of Cooling and Forced Convection Correlations (Dittus-Boelter, Sieder-Tate) Boundary Layer Theory in Convective Heat Transfer