THE ART OF PHYSICS AN INTERACTIVE OBSERVATORY

The Foucault
Pendulum

Watch the Earth turn beneath a swinging pendulum. Discover the invisible forces that connect a single brass sphere to the motion of our planet.

51° 31′ N / THE ORIGINAL EXPERIMENT / 1851
01 / INTERACTIVE EXPERIMENTREAL-TIME PHYSICS ENGINE

A world in motion.

Interact with the instrument. Change its environment. Observe the laws of nature unfold.

LIVE SIMULATIONINITIALISING 3D
◌  DRAG TO ORBIT  /  SCROLL TO ZOOM— FPS
LIVE TELEMETRYAPPROXIMATE MODEL VALUES
Elapsed time0.0 s
Swing period—
Bob speed—
Mechanical energy—
Natural drift—
Boosted drift—
Horizontal offset—
Current tilt—
OBJECT INSPECTOR · CLICK A COMPONENT

Explore the observatory

Click the pendulum bob, suspension cable, pivot, or compass dial to learn about its purpose.

EXPERIMENT 01

Set latitude to 0° and watch the predicted natural drift vanish. Then try +90° and −90° to compare both poles.

THE DISCOVERY

About This 3D Simulation

A Foucault pendulum is a long, freely swinging weight that reveals the rotation of Earth without relying on a telescope, a compass needle, or the apparent movement of the stars. When the pendulum starts moving, its plane of oscillation tends to keep its orientation in inertial space. Meanwhile, the building, the floor, and the observer rotate with Earth. Over time, a stationary observer sees the swing direction turn relative to the floor. This elegant demonstration, first made famous by Léon Foucault in Paris in 1851, connects a familiar swinging object to the motion of an entire planet.

This interactive observatory places a luminous brass bob beneath an open architectural dome. An adjustable suspension cable meets an illustrated azimuth dial on the floor, where a cyan projected trace makes the oscillation visible. The pavilion's columns, radial ceiling trusses, overhead winch, and ornamental details are procedurally assembled in Three.js. You can orbit the camera, inspect the brass bob and support mechanism, switch to an overhead view, and watch a reference line indicate the predicted rotation of the swing plane.

Change the geographic latitude to compare the equator, mid-latitudes, and the poles. Increase the length to see the swing period grow; raise the initial release angle to give the bob a wider path. Add damping to represent gradual loss of motion through air resistance and friction. The Earth-rotation multiplier makes the very slow real-world precession visible in a short lesson: it is deliberately an educational exaggeration, not a claim that Earth is actually spinning faster. A separate real-Earth preset restores the natural rotation rate.

The numerical model uses a horizontal small-angle oscillator with a Coriolis acceleration. It is useful for understanding latitude, period, energy, and a changing swing plane, but it does not simulate every feature of a full spherical pendulum. Read the explanation, inspect the live measurements, and try the suggested experiments to connect the visual demonstration with quantitative physics.

THE SCIENCE

How It Works

The starting idea: an inertial swing plane

In an idealised setup, a pendulum released without sideways motion swings in a nearly fixed vertical plane. A terrestrial observer is attached to a planet that turns once relative to the distant stars in roughly 23 hours 56 minutes. Because the floor changes orientation underneath the swing, the trace appears to rotate. The local effect depends on the vertical component of Earth’s angular-velocity vector, which is proportional to the sine of latitude.

The natural precession rate is Ωₚ = Ω⊕ sin(φ), where Ω⊕ ≈ 7.292 × 10⁻⁵ rad/s is Earth’s sidereal rotation rate and φ is geographic latitude. In degrees per hour, this is approximately 15.04° × sin(φ). At the North Pole the plane makes a full turn relative to the floor in a sidereal day; at the equator the ideal local precession rate is zero. The direction reverses between hemispheres.

Components, motion and the numerical model

The brass bob supplies inertia, the long cable constrains it to remain approximately a fixed distance from the upper pivot, and gravity restores the bob toward its lowest point. The floor dial supplies the Earth-fixed reference. The displayed trajectory is the horizontal projection of the calculated motion, not a physical mark made by the bob. Architectural columns, dome ribs and the decorative winch provide context and do not affect the dynamics.

For small deflections, the horizontal east and north coordinates (x, z) obey ẍ = −(g/L)x + fż − cẋ and z̈ = −(g/L)z − fẋ − cż. Here g = 9.80665 m/s², L is cable length, c is a damping coefficient, and f = 2Ω⊕ sin(φ) times the demonstration multiplier. These coupled differential equations include gravity, a simplified linear damping force, and the horizontal Coriolis deflection. They are integrated numerically with fourth-order Runge–Kutta steps.

Oscillation period and energy

The familiar small-amplitude period is T ≈ 2π√(L/g). A 14 m pendulum therefore takes roughly 7.51 seconds for a complete oscillation. It is mainly the length, not the bob’s mass, that sets the period in this approximation. The bob starts from rest at the selected release angle; changing either length or release angle resets the experiment so that its initial conditions remain meaningful.

Live kinetic energy is calculated as ½m(vₓ² + v_z²). Potential energy uses mg[L − √(L² − x² − z²)], consistent with the pendulum’s spherical cable constraint for the displayed height. Their sum is an educational energy estimate. If damping is zero, the Coriolis force does no work, although small numerical and modelling differences may remain. Increasing damping gradually reduces motion and energy.

Controls, interpretation and limits

Use the latitude slider to change the north–south location, the length slider to change the period, and the amplitude slider to choose the initial displacement. Playback speed advances simulation time faster; the Earth-rate boost instead magnifies only the Coriolis effect so precession becomes visible. The natural drift readout always reports the actual theoretical Earth-rate prediction, while the boosted readout is explicitly artificial. Camera presets change only the viewing angle.

This model assumes modest angles, a rigid massless cable, no wind, no pivot asymmetry, and a locally flat Earth-fixed horizontal reference frame. It does not resolve spherical-pendulum nonlinearities, detailed suspension stresses, or real atmospheric turbulence. At very high Earth-rate boosts the altered Coriolis term is a hypothetical system rather than a faithful accelerated recording of a real installation. Compare settings critically: the best scientific insight comes from connecting observed motion with the formulas and stated assumptions.

FOUCAULT PRECESSIONΩp = Ω⊕ sin(φ)

At +52.2° latitude, the ideal natural drift is approximately 11.9° per hour.

FIELD MANUAL

How to Use This Simulation

01

Navigate the instrument

Drag inside the viewer to orbit, scroll or pinch to zoom, and pan with the right mouse button. Use Top view to compare the projected path to the azimuth dial. Reset view returns to the opening camera.

02

Run and inspect

Select Pause to stop simulation time without changing the camera; Resume continues the motion. Reset starts a new swing. Click the brass bob, cable, pivot, or floor dial to inspect an individual component.

03

Adjust the physics

Move the latitude, damping, Earth-rate and playback sliders while running. Changing cable length or release angle starts a fresh experiment. Switch the projected trail and idealised swing-plane guide on or off.

04

Read the evidence

Compare oscillation period, horizontal offset and mechanical energy in the live panel. Distinguish natural drift from the deliberately boosted rate. Try one parameter at a time, then compare your observation with the equation above.

FURTHER QUESTIONS

Frequently Asked Questions

Ten detailed explanations of the physics, mechanics and limitations behind the experiment.

01What does a Foucault pendulum actually prove?

A Foucault pendulum provides a direct mechanical demonstration that the ground beneath it rotates. A carefully released, sufficiently long pendulum tends to keep its swing plane oriented relative to inertial space. The laboratory floor moves with Earth, so the swing direction appears to drift relative to marked directions on that floor. The effect is not an invisible motor turning the pendulum; it arises because the observer is inside a rotating reference frame. In this simulation the floor dial is the Earth-fixed reference and the brass bob is the moving oscillator. It makes the central argument visible, although the teaching model is simpler than a full research-grade apparatus.

02Why does the precession rate depend on latitude?

Only the local vertical component of Earth’s angular velocity makes the swing plane turn relative to the horizontal floor in the ideal Foucault analysis. That component equals Earth’s angular speed multiplied by the sine of geographic latitude. It is greatest in magnitude at the poles and zero at the equator. At a northern latitude the observed orientation rotates in one direction; at the matching southern latitude it rotates in the opposite direction. A setting near 52° north gives a natural drift of roughly twelve degrees per hour. Try moving the latitude slider from zero to ninety degrees while keeping the cable length and release angle unchanged to isolate the latitude effect.

03What are the functions of the bob, cable, pivot and floor dial?

The bob is a concentrated mass that makes the pendulum’s motion easy to follow and keeps a useful amount of inertia. The cable fixes the distance between the bob and the support point, while an ideal low-friction pivot lets the pendulum swing in more than one vertical plane. A real exhibit must also minimise vibration, air currents, and asymmetric support forces. The compass-style floor dial serves a different purpose: it provides marks that rotate with the building, so viewers can compare the swinging motion to their Earth-bound surroundings. In the 3D scene, clicking each major component opens a short explanation. The dome itself is decorative and is not part of the physics equations.

04How does the Coriolis effect appear in the mathematical model?

The Coriolis effect appears when motion is described in a coordinate system attached to a rotating body such as Earth. For horizontal velocities, the simplified acceleration is perpendicular to the velocity and has a coefficient f = 2Ω⊕ sin(φ), sometimes called the Coriolis parameter. The model couples eastward and northward motion through terms proportional to the other direction’s velocity. It also includes gravitational restoring acceleration and optional linear damping. A perpendicular Coriolis force changes the direction of velocity without directly adding mechanical energy. The animation numerically integrates these equations in small time steps. The boost control scales the Coriolis coefficient for teaching and therefore does not reproduce the exact dynamics of a real, unboosted pendulum.

05How does pendulum length change its period?

For a small swing angle, the period of a simple pendulum is approximately T = 2π√(L/g), where L is the effective length and g is local gravitational acceleration. Doubling the length increases the period by a factor of √2 rather than doubling it. In this observatory, moving the length slider between ten and eighteen metres updates the projected period and the suspension geometry. The pendulum is released again when you change length, so the initial amplitude remains well defined. The simple formula omits finite-angle corrections, cable flexibility, and nonuniform gravity. It also shows why replacing the brass bob with a different mass would not substantially alter the ideal small-angle period.

06How quickly would this pendulum precess in a real museum?

Real-world Foucault precession is usually much slower than the motion shown in accelerated educational demonstrations. Its ideal rate is approximately 15.04 degrees per hour multiplied by the sine of the local latitude. At 52.2° north, that works out to about 11.9 degrees per hour, which means a noticeable change in floor alignment takes many minutes. A complete circuit there takes roughly thirty hours, while a polar pendulum completes its relative rotation in one sidereal day. The live panel separates this natural rate from the artificially boosted teaching rate. Press the real-Earth preset to remove the exaggeration, then compare the slow change with the faster demonstration setting.

07How should I use the controls to run a useful experiment?

Begin with the demo preset, an initial amplitude around ten degrees, and a camera looking down toward the floor dial. Start the simulation and note the apparent swing direction relative to the fixed compass marks. Pause to examine the projected trail, then change only latitude and repeat. Try the equator and a pole as limiting cases. For a second experiment, keep latitude fixed but increase cable length: the period readout grows, while the ideal natural precession rate stays controlled by latitude. Finally, turn damping up to see energy decrease. Reset returns the bob to its configured initial release. Change one parameter at a time whenever you want to identify a clear cause-and-effect relationship.

08Why does the motion gradually fade when damping is enabled?

A real pendulum loses energy to air drag, tiny vibrations, deformation, and friction at the support. These dissipative effects eventually stop the motion unless a museum mechanism gently maintains the oscillation. This simulation uses a simplified linear damping acceleration that is opposite to horizontal velocity. Larger damping values remove kinetic and potential energy more rapidly, making the swing envelope shrink. The natural Coriolis term itself does not dissipate energy because its acceleration is perpendicular to the horizontal velocity. If you set damping to zero, the model should oscillate for much longer, aside from small integration errors. The demonstration intentionally omits detailed aerodynamic drag laws and active driving mechanisms.

09What does this simulation leave out compared with a real Foucault pendulum?

The displayed motion uses a linear small-angle approximation with two horizontal coordinates and an idealised cable. Real pendulums follow a three-dimensional spherical constraint and may display elliptical motion, suspension imperfections, nonlinear amplitude corrections, and effects from air currents or subtle asymmetries. The rendered support, dome, and metallic surfaces are illustrative geometry rather than a structural-engineering calculation. The energy display is also approximate, because it uses horizontal velocity components and an inferred vertical rise. Large demonstration multipliers artificially increase the Coriolis force and should not be interpreted as real Earth conditions. These omissions are explained so visitors can distinguish a useful physical teaching model from an experimentally calibrated digital twin.

10Why can the experiment reveal Earth's motion without watching the sky?

The key is that an oscillator with very little unwanted sideways force provides a reference orientation that does not simply follow the rotation of the floor. An observer does not need to see the stars: they compare the plane of the pendulum’s motion with marks fixed to their own building over a period of time. The latitude-dependent precession agrees with what a rotating terrestrial coordinate system predicts. A well-designed real installation must account for disturbances that might mimic a slow drift, which is why long cables, heavy bobs, careful release procedures, and low-friction pivots matter. The exercise illustrates an important theme in physics: measurements inside a rotating frame can reveal the frame’s motion.

Made with 🤖 ChatGPT · Three.js