Froude Number or Reynolds Number for Scale Model Similarity in Hydraulic Engineering

When civil engineers build physical scale models of hydraulic structures, dams, channels, or coastal defenses, they face a fundamental question: which dimensionless number governs the similarity between the model and the full-scale prototype? The two primary candidates are the Froude number, which captures the ratio of inertial to gravitational forces, and the Reynolds number, which captures the ratio of inertial to viscous forces. Selecting the wrong one leads to model data that cannot be reliably scaled up to predict real-world performance. This article explains the theoretical basis of each number, the practical constraints that force engineers to choose, and the decision framework used in hydraulic laboratories worldwide. Before constructing any hydraulic structure, engineers must also navigate regulatory requirements such as who should apply for a building permit understanding the distinction between owner and contractor obligations, which parallels the case-by-case judgment needed in similitude analysis.

Understanding Dimensional Analysis in Hydraulic Modeling

The Principle of Similitude

Similitude is the science of establishing a reliable relationship between a scale model and its prototype. Three types of similarity must be satisfied for a model to produce meaningful results:

1. Geometric similarity – The model is an exact scaled replica of the prototype in shape. All linear dimensions are scaled by the same factor.
2. Kinematic similarity – The flow patterns (velocity fields, streamlines, and accelerations) are proportionally identical between model and prototype.
3. Dynamic similarity – The ratios of forces acting on fluid particles in the model match those in the prototype.

Dynamic similarity is the most challenging to achieve because multiple forces gravity, viscosity, surface tension, elasticity, and pressure act simultaneously on a fluid. Engineers must identify the dominant force and select the appropriate similarity criterion. A dimensionless number is a pure ratio of forces that has no units. If this number is equal for both model and prototype, the relative importance of the underlying physical effects is preserved. The two most important dimensionless numbers in hydraulic modeling are the Froude number and the Reynolds number:

Where V is characteristic velocity, L is characteristic length, g is gravitational acceleration, ρ is fluid density, and μ is dynamic viscosity. Understanding these distinctions is essential when determining who should apply for a building permit owner versus contractor responsibilities in hydraulic structure projects.

When to Use the Froude Number

Free Surface Flows and Gravitational Dominance

The Froude number is the correct similarity criterion whenever gravitational forces dominate the flow behavior. This occurs in nearly all open channel flows, where a free surface exists and the water surface profile is determined by the balance between gravity and inertia. The source article from Engineering Civil Portal states plainly: Froude number is used when gravitational forces are predominant in the channel flow. Specific applications include:

• Weirs and spillways – Flow over dam spillways is governed by the fall of water under gravity. Froude similarity ensures the nappe shape, discharge coefficient, and energy dissipation patterns scale correctly.
• Hydraulic jumps – The transition from supercritical to subcritical flow produces a hydraulic jump whose sequent depth ratio depends directly on the upstream Froude number.
• Ship and vessel hydrodynamics – Wave resistance and hull performance in calm water are Froude-dominated because wave generation is a gravitational phenomenon.
• Coastal and harbor models – Wave propagation, refraction, diffraction, and run-up on beaches are all gravity-driven processes.
• River and floodplain modeling – Stage-discharge relationships and flood wave propagation in natural channels follow Froude similarity.

Practical Scaling Under Froude Similarity

When the Froude number is held constant between model and prototype, the velocity scale follows directly from the geometric length scale. If the model length ratio is Lr = Lm / Lp, then the velocity ratio Vr = √Lr, the time ratio Tr = √Lr, and the discharge ratio Qr = Lr^5/2. This straightforward scaling is one reason Froude models are widely used. An engineer working with a 1:25 scale model of a spillway knows that velocities in the model will be one-fifth of prototype velocities, and discharge will scale by 1/3125.

Limitations of Froude Modeling

The primary drawback is that the Reynolds number is not preserved. In a small-scale model, the flow may become laminar or transitional when the prototype flow is fully turbulent. This mismatch can affect friction factors and boundary layer development. Engineers must compensate by roughening the model surface or accepting a restricted range of validity. When examining structural details of hydraulic systems, should I glue screws thread locking guide decisions become important when constructing flume walls and instrumentation mounts in scale models.

When to Use the Reynolds Number

Viscous Dominance in Confined Flows

The Reynolds number is the governing similarity parameter when viscous forces dominate the flow behavior. The source article notes that Reynolds number is adopted when viscous forces are predominant in the channel flow. This occurs in pressurized pipe systems, boundary layer flows, and any situation where the flow is fully confined. Key applications include:

• Pipe flow systems – Head loss due to friction in pipes depends on the Reynolds number and relative roughness. Reynolds similarity ensures that the Darcy-Weisbach friction factor scales correctly.
• Wind tunnel testing – Aerodynamic drag on bridges and buildings is primarily a viscous phenomenon. Reynolds number matching ensures that boundary layer separation points are reproduced.
• Sediment transport – The initiation of sediment motion depends on the balance between fluid shear and particle weight, making Reynolds number relevant at the grain scale.
• Flow through porous media – Darcy flow and the transition to Forchheimer flow are governed by pore Reynolds numbers.

The Impracticality of Simultaneous Similarity

The source article makes a critical observation: it is almost impossible to make the Froude number and Reynolds number identical in model and prototype. The fundamental conflict is that the two dimensionless numbers impose opposing scaling laws on velocity:

At a 1:20 geometric scale, Froude similarity demands model velocities roughly one-fifth of prototype, while Reynolds similarity demands velocities twenty times larger. These are irreconcilable with the same fluid. Using a different fluid with a much lower viscosity can help, but practical limits exist. Mercury has a low kinematic viscosity but is hazardous and expensive. This constraint forces engineers to choose the dominant force and accept that the other dimensionless number will differ.

Practical Scaling Under Reynolds Similarity

Under Reynolds similarity, the velocity scale is inversely proportional to the length scale: Vr = 1/Lr, meaning a 1:10 scale model requires velocities ten times larger than the prototype. In a water model, such high velocities demand unrealistic pump capacities and may cause cavitation. In a wind tunnel, the velocity increase is feasible but power requirements grow rapidly.

Decision Framework and Practical Considerations

Step-by-Step Selection Process

The source article emphasizes that the use of these numbers should be judged case by case. Experienced hydraulic modelers follow a structured approach:

1. Identify the dominant forces – Determine whether gravitational forces (free surface, waves, weir flow) or viscous forces (pipe flow, boundary layer, drag) govern the phenomenon.
2. Check for a free surface – Any flow with an air-water interface is almost always Froude-dominated. Flows without a free surface are Reynolds-dominated.
3. Assess the Reynolds number range – Even in Froude-dominated flows, ensure the model Reynolds number remains above the turbulent threshold (Re > 2000). If not, consider surface roughening or a larger scale.
4. Consider secondary effects – Surface tension (Weber number), elasticity (Mach number), and cavitation (Euler number) may become relevant in specialized cases.
5. Validate with field data – Calibrate the model against prototype measurements when possible to confirm the chosen similarity criterion.

Case Study: Open Channel Flow Modeling

Consider a laboratory building a 1:20 scale model of a river reach to study flood conveyance. The river is 50 m wide and 4 m deep at bankfull discharge of 400 m³/s. The prototype flow has Re ~ 10⁷ (fully turbulent) and Fr ~ 0.3 (subcritical, gravity-dominated). The correct choice is Froude similarity. The model discharge is Qp × Lr^5/2 = 400 / 1789 = 0.224 m³/s, or 224 L/s. The model Reynolds number is 10⁷ / 89.4 = 112,000, still fully turbulent. If the scale were reduced to 1:100, the model Re would drop to 10,000, approaching the transitional regime. Understanding these scaling decisions is as important as knowing rigid foam sheathing placement should you insulate inside or outside the framing when designing hydraulic laboratory structures.

Scale Effects and Their Mitigation

Every physical model suffers from scale effects deviations caused by imperfect similarity:

• Viscous scale effects – In Froude models, the lower Reynolds number increases the relative importance of viscosity, altering friction factors and boundary layer development.
• Surface tension effects – At very small scales, capillary forces (Weber number) can distort free surface shapes in weir models and wave tanks.
• Roughness scaling – Model roughness elements must be scaled proportionally, but manufactured surfaces may not match natural hydraulic roughness.
• Sediment transport distortion – In mobile bed models, both Froude and Reynolds similarity apply simultaneously, requiring calibration against empirical formulas.

Physical versus Numerical Models

Computational fluid dynamics has not eliminated the need for physical scale models. Physical models remain indispensable when flow geometry is too complex for reliable mesh generation, turbulence models cannot resolve relevant scales (air entrainment on spillways, vortex formation), or validation data for numerical codes is needed. The optimal approach combines both methods: physical models reveal unexpected phenomena and provide validation data, while numerical models extend the parameter space.

Best Practices Summary

• Identify the dominant force mechanism before selecting the similarity criterion. For free surface flows, use Froude similarity. For confined viscous flows, use Reynolds similarity.
• Check the model Reynolds number even in Froude-scaled models to ensure turbulent conditions are maintained.
• Document expected scale effects and report them alongside model results.
• Use the largest practical scale factor to minimize mismatch in the secondary dimensionless number.
• Validate against field measurements or published data when available.
• For long river reaches, consider distorted models with a distortion ratio limited to 5:1.

The choice between Froude number and Reynolds number in scale models is not a matter of preference but a rigorous engineering decision based on the physics of the flow. The source article concludes what every hydraulic modeler knows: the case-by-case evaluation guided by engineering judgment remains the most critical tool in the modeler’s arsenal. A well-designed scale model, built on the correct similarity foundation, delivers predictions that save millions in construction costs and prevent failures in full-scale hydraulic infrastructure.