Skip to content

Froude Number

A dimensionless speed–gravity–length comparison used to assess gravity-mediated motion and bounded dynamic similarity.

Core Idea

The Froude number compares a characteristic speed with a speed scale set by gravity and a characteristic length. In much of fluid engineering it is written \(Fr=v/\sqrt{gL}\), where \(v\) is speed, \(g\) gravitational acceleration and \(L\) a declared length. The units cancel, so systems of different absolute size can be compared in terms of how strongly motion competes with gravity over that length.[1][2]

The ratio is informative only after its regime, length and convention are specified. In an open channel, \(L\) is often hydraulic depth and \(Fr=1\) marks critical flow in the stated shallow-gravity-wave model. In ship testing, waterline length and water depth yield different Froude numbers. In locomotion, leg length is used, and some authors call the squared ratio \(v^2/(gL)\) the Froude number; the fluid-engineering value is its square root. Equal values of a properly matched definition are useful evidence for one aspect of dynamical similarity, not proof that resistance, wave pattern or gait is identical.[1][2][3]

Structural Signature

Sig role-phrases: moving system and speed — gravitational acceleration — declared characteristic length — dimensionless convention — regime-bounded interpretation.

  • Moving system and speed. Mean channel velocity, vessel speed through water or forward gait speed supplies \(v\). Without a motion variable this is only a possible gravity-length scale.[1][2][3]
  • Gravitational acceleration. \(g\) supplies the restoring or weight-related acceleration against which motion is compared. Substituting viscosity or sound speed defines another parameter, not Froude number.
  • Declared characteristic length. Hydraulic depth \(D=A/T\) in a channel, waterline length in a ship test, and leg length in a gait comparison serve different physical questions. An unexplained \(L\) makes numerical comparison ambiguous.[1][2][3]
  • Dimensionless convention. The common engineering form is \(v/\sqrt{gL}\). A biomechanics source may use \(v^2/(gL)\) and call the first form dimensionless speed. For nonnegative speed these order cases alike, but reported numbers and threshold values differ.[2][3]
  • Regime-bounded interpretation. The ratio isolates gravity-relative motion only within a model. Channel criticality, ship-wave scaling and gait comparison require different outcome checks. Friction, geometric form, stiffness and landing dynamics remain outside this single scalar.[1][2][4]

Removing either gravity or a declared length destroys the ratio's physical identity. Removing the outcome model does not make the arithmetic impossible, but it does remove any warranted claim about a transition or successful scale prediction.

What It Is Not

Froude number is not hull speed, which names a particular length-related ship-wave speed estimate. A vessel can have a length Froude number at any speed; there is no universal hard speed barrier encoded in the parameter. The frozen Hull Speed Wikipedia page was the discovery candidate from which this entry was reframed, not evidence that the two names are synonymous.[2]

It is not Reynolds number. The latter tracks inertia relative to viscosity and is separately recorded in the ITTC model-resistance procedure. Matching a model ship's Fr cannot, by itself, guarantee equal viscous friction at full scale. Froude number also is not a universal law of exact dynamic similarity. A gait study explicitly finds equal Fr necessary but insufficient for the similarity it analyzes when leg stiffness and vertical landing speed vary.[2][4]

Scope of Application

Open-channel hydraulics uses hydraulic depth \(D=A/T\), cross-sectional area divided by water-surface top width. A USGS derivation relates \(Fr=v/\sqrt{gD}\) to specific-energy criticality: \(Fr=1\) at critical flow, where mean velocity equals the shallow-water gravity-wave speed in that model. Subcritical and supercritical flow are meaningful regime labels there; they should not be transferred unchanged to animal gaits or all ship-wave behavior.[1]

Naval testing uses length Fr, normally with waterline length, while also tracking depth Fr when the tank or water depth matters. ITTC's original procedure records Reynolds number and frictional-resistance treatment separately, and states that its resistance-test procedure covers model scale rather than full-scale extrapolation. These qualifications are why the Froude ratio is a controlled similarity coordinate, not a complete vessel prediction by itself.[2]

Comparative biomechanics scales forward speed by gravity and leg length. Alexander and Jayes' original abstract reports useful gait correspondences among cursorial quadrupeds at matched Fr but weaker agreement across less alike animals. A later theoretical gait analysis uses the squared definition and concludes that matching it alone is insufficient; stiffness and landing conditions can differ. These sources support bounded cross-size comparison, not a universal walk-to-run number.[5][4]

Clarity

The expression \(v/\sqrt{gL}\) can be read as actual speed divided by a gravity-based speed scale. For a channel, say which section and use \(L=D=A/T\); for a ship, distinguish waterline length from water depth; for an animal, state how leg length was measured. Two systems with the same numerical Fr under different length definitions have not thereby matched the same physical comparison.[1][2]

When a paper reports \(Fr=0.5\) in gait mechanics, check whether its author means the squared \(v^2/(gL)\) or the square-root speed \(v/\sqrt{gL}\). The first corresponds to approximately $0.707$ on the second scale. The transformation preserves ordering for nonnegative speed but not literal threshold numbers. A claim that a single threshold transfers among channels, ships and animals would compound two errors: convention mismatch and regime mismatch.[3]

Manages Complexity

The parameter compresses three dimensional inputs into one coordinate. In hydraulics that makes the relation between flow speed and gravity-wave speed immediately visible. In ship tests, it helps choose model speeds appropriate for gravity-mediated wave behavior rather than trying to compare vessels at the same absolute meters per second. In gait research, leg-length scaling makes a small and a large animal's forward speeds comparable under a gravity-related hypothesis.[1][2][5]

That compression has a price. Ship friction also depends on viscosity and surface condition; channel jumps and loss depend on geometry and boundary conditions; gait timing depends on limb stiffness and landing kinematics. A useful Froude comparison is therefore an entry point to analysis, followed by checks of the other forces and structures relevant to the outcome.[2][4]

Abstract Reasoning

First identify the gravity-mediated process: shallow free-surface waves, wave-making around a hull, or weight-supported leg motion. Then choose a speed and a characteristic length that actually participate in that process, and declare whether Fr means \(v/\sqrt{gL}\) or \(v^2/(gL)\). Verify dimensional cancellation, calculate the ratio and compare only like conventions. Finally ask what the ratio is meant to predict and what independent parameters the proposed prediction ignores.[1][2][3]

For instance, a ship model and prototype can be arranged to have the same length Fr, but the ITTC analysis still separates viscous/frictional effects. Conversely, two animals at the same gait Fr can remain mechanically unlike if dimensionless leg stiffness differs. The counterfactual test is not “Would equal Fr make everything the same?” but “Which gravity-related behavior would change if speed were increased relative to \(\sqrt{gL}\) while the relevant other conditions were controlled?”[2][4]

Knowledge Transfer

The literal transfer from hydraulic channels to naval models and gait studies is a method of dimensional comparison: select a physically justified \(L\), normalize speed by a gravity scale, then test the expected similarity rather than assuming it. All three share speed, gravitational acceleration, length and an interpreted ratio. They do not share a single universal threshold, length choice or set of other governing forces.[1][2][5]

This also explains why cross-domain reach does not make Froude Number a prime abstraction. Its essential \(g\), \(L\) and velocity roles remain physical and gravity-specific. A general prime of normalization or similarity might be instantiated by the measure, but replacing those roles with metaphorical “organizational gravity” would no longer be the named quantity.[3]

Examples

Critical open-channel section. Let a channel section have wetted cross-sectional area \(A\), top width \(T\) and mean velocity \(v\). The hydraulic depth is \(D=A/T\), giving \(Fr=v/\sqrt{gD}\). USGS derives \(Fr=1\) at critical flow under its specific-energy model: the fluid's mean speed equals the small gravity-wave celerity. A changed width or depth changes the normalizing length, not just the numerator.[1] Mapped back: speed = section mean \(v\); gravity = \(g\); length = \(A/T\); convention = square-root form; interpretation = channel criticality, not all-purpose similarity.

Towing-tank vessel. ITTC defines length \(Fr=v/\sqrt{gL}\) with \(L\) normally waterline length, and a separate depth Froude number with water depth. A model test can choose a speed corresponding to a target length Fr, then still measure resistance and treat Reynolds/friction and tank effects separately. A matched Fr is a bounded control of gravity/wave scaling, not a prediction that total model and full-scale drag coefficients coincide.[2] Mapped back: speed = model velocity through water; gravity = \(g\); lengths = waterline length or tank depth; convention = square-root forms; interpretation = relevant wave/depth scaling with separate correction obligations.

Mammal gait comparison. Alexander and Jayes test whether different-sized mammals show comparable features of gait at matched Froude number. Their abstract reports stronger performance among cursorial quadrupeds than across very different animals. Later gait work writes \(Fr=v^2/(gL_{leg})\) and shows that equal values still do not guarantee similarity when stiffness or vertical landing speed changes.[5][4] Mapped back: speed = forward locomotion; gravity = \(g\); length = leg length; convention = explicitly squared in the later paper; interpretation = a hypothesis for some gait features, subject to additional biomechanical variables.

The hydraulic and gait examples are deliberately unlike. Their common structure is the normalized speed–gravity–length relation, while their thresholds, geometries and supported predictions differ.

Structural Tensions

Common normalization versus regime-faithful length. A standard, easy-to-report characteristic length makes comparisons compact. But the length that actually governs an outcome can be hydraulic depth, waterline length, water depth or a functional leg length; replacing it with a convenient standard can obscure the physics. Choosing a carefully tailored length improves local fidelity but makes unqualified cross-study comparison harder. Diagnostic: Does the length selected for a convenient comparison actually control the gravity-mediated behavior being tested?[1][2][3]

Scalar parsimony versus complete dynamic account. Matching Fr gives a tractable first comparison and can reduce a scaling problem to one coordinate. Yet equal Fr leaves Reynolds effects, hull geometry, limb stiffness and landing motion free to differ. Requiring all relevant parameters improves the prospective prediction but costs measurement effort and sacrifices the simplicity of a one-number screen. Diagnostic: Which omitted quantity could falsify the proposed similarity, and is the extra measurement needed for the decision at hand?[2][4]

Structural–Framed Character

Froude Number has a strongly structural physical core but requires domain framing to be useful. Evaluative weight is absent from the number itself: a high Fr is not intrinsically better or worse. Human-practice dependence enters through the selected length, whether a square-root or squared convention is used, and which comparison matters. Institutional origin is reflected in technical procedures such as ITTC's model-test reporting, though the dimensionless relation is not created by an institution. Vocabulary travel is real across hydrodynamics and biomechanics, but import versus recognition requires a genuine speed–gravity–length relation, not an analogy. The same formula family can support different local claims without entailing a single cross-domain regime boundary.[2][3]

Its character: a gravity/inertia similarity parameter with a stable dimensional relation and setting-dependent interpretive strength.

Structural Core vs. Domain Accent

The core is a dimensionless comparison of motion speed to a gravity-and-length scale, with the convention stated. The accents determine what the comparison means. Hydraulic depth and water-wave celerity make the \(Fr=1\) open-channel result possible; waterline length makes a vessel-wave coordinate; leg length makes a gait comparison. None of those length choices is arbitrary decoration. The same numerical value using different lengths can address different questions.[1][2][3]

The boundary between core and accent also prevents a false prime promotion. If gravity or characteristic length is replaced by a generic “influence,” the identity dissolves into general normalization. Whether such a portable normalization pattern merits a separate prime is an unadmitted future-prime question; no prime parent is asserted from that suggestion. The actual strict parent proposed here is domain-specific Dimensionless Quantity, whose dimension-one genus does not erase this entry's gravity-specific differentia.

This entry is a kind of Dimensionless Quantity.

Dimensionless Quantity is the broader abstraction this entry instantiates: the Froude number is a quantity of dimension one whose numerical value persists under coherent changes of base units. Physical quantity is a higher ancestor through that parent. Reynolds Number is a useful sibling comparison in model testing, not a Froude synonym or parent. Hull speed is the frozen candidate's narrower ship-wave topic, not the broader abstraction in this graph. No cross-domain prime edge is asserted merely because the normalized ratio travels between engineering and biology.

Relationships to Other Abstractions

Local relationship map for Froude NumberParents appear above the current abstraction, mutual partners to the right, and children below. Node labels state whether each abstraction is prime or domain-specific; colors identify relation types.Froude NumberDOMAINDomain-specific abstraction: Dimensionless Quantity — is a kind ofDimensionlessQuantityDOMAIN

Current abstraction Froude Number Domain-specific

Parents (1) — more general patterns this builds on

  • Froude Number is a kind of Dimensionless Quantity Domain-specific

    Froude number is a dimensionless physical quantity built from speed, gravity and characteristic length.

Hierarchy path (1) — routes to 1 parentless root

Neighborhood in Abstraction Space

Froude Number sits in a sparse region of the domain-specific corpus (86th percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Classical Mechanics & Orbital Kinematics (12 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-08

Not to Be Confused With

  • Hull speed: a particular estimated ship-wave speed, not the dimensionless Froude parameter at every speed.[2]
  • Reynolds number: viscosity-relative flow parameter; matching one does not automatically match the other.[2]
  • An absolute speed or speed/length ratio with units: dimensional cancellation requires \(g\) and a specified length in the displayed convention.[2]
  • Universal transition at Fr=1: directly supported here for open-channel criticality under a particular hydraulic-depth model, not for all vessels or gaits.[1]
  • Complete dynamic similarity: equality of one ratio need not align friction, geometry or limb mechanics.[2][4]

References

[1] U.S. Geological Survey, Open-File Report 88-707, ch.11 p.72, equations (11-3)–(11-7), original citation (PDF retrieval failed in independent review). The formula, hydraulic-depth definition, critical-flow relation and wave-celerity condition were independently checked against the U.S. Bureau of Reclamation Water Measurement Manual, ch.2 §§11–12, equations (2-24)–(2-31). registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n

[2] International Towing Tank Conference, Recommended Procedure 7.5-02-02-01: Resistance Test (2008), PDF p.2 definitions of length/depth Froude and Reynolds numbers; pp.9–10 separate resistance analysis and corrections. This procedure explicitly covers model-scale testing only. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j ↩k ↩l ↩m ↩n ↩o ↩p ↩q ↩r ↩s ↩t ↩u ↩v ↩w ↩x

[3] Monica A. Daley and Aleksandra Birn-Jeffery, “Scaling of Avian Bipedal Locomotion Reveals Independent Effects of Body Mass and Leg Posture on Gait”, Journal of Experimental Biology 221 (2018), original article definition of squared gait Froude number and its square-root dimensionless speed. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h ↩i ↩j

[4] S. R. Bullimore and J. M. Donelan, “Criteria for Dynamic Similarity in Bouncing Gaits”, Journal of Theoretical Biology 250 (2008), original paper abstract and introduction, explicitly states equal Fr is necessary but not sufficient for its bouncing-gait model and describes leg-stiffness and landing-speed conditions. registry ↩a ↩b ↩c ↩d ↩e ↩f ↩g ↩h

[5] R. McN. Alexander and A. S. Jayes, “A Dynamic Similarity Hypothesis for the Gaits of Quadrupedal Mammals”, Journal of Zoology 201 (1983), publisher abstract; full paper not independently open in this staged review. registry ↩a ↩b ↩c ↩d