This tutorial visualizes how light is guided in an optical fiber. The fiber path is drawn as a draggable spline curve. The core and cladding follow the same curved path, and the light rays propagate along the core while reflecting from the core-cladding boundary.

Mathematical Foundation

A step-index optical fiber has a high-index core and a lower-index cladding (n_core > n_clad). Light is guided when it enters within the fiber acceptance cone, whose size is set by the numerical aperture and maximum acceptance angle:

The simulator compares the configured launch angle with theta_max. If the launch angle is too large, the ray is shown escaping because it is outside the acceptance cone. If it is guided, the reflection spacing is calculated from the refracted angle inside the core:

So changing n_core changes the internal ray angle. Changing n_clad changes NA and theta_max, so it changes how close the launch condition is to cutoff. For visualization, the simulator uses this acceptance margin to slightly change the reflected segment spacing, making both n_core and n_clad visible in the path.

Total Internal Reflection

Inside the core, guidance is caused by total internal reflection at the core-cladding boundary. The critical angle there is:

When the incidence angle at the boundary exceeds this critical angle, the ray remains in the core. (If the cladding is physically absent, this simplified simulator treats the fiber as not guided because the controlled lower-index boundary is missing.) In the visualization, guided light is traced as straight global line segments; each segment intersects the curved core boundary, then reflects using the local boundary normal so the incident and reflected angles are equal.

Fiber Shape And Bend Loss

The fiber path is a Catmull-Rom spline through draggable control points; dragging a point changes the bend radius. Tight bends can cause radiation loss because the guided mode cannot remain fully confined. The simulator estimates bend loss from the minimum bend radius along the spline — a simplified planning-style estimate, not a full electromagnetic mode solver.

Single Source And WDM

A single optical source uses one wavelength. WDM (wavelength division multiplexing) injects multiple wavelengths into the same fiber; the simulator can show one channel or several channels using different colors.

Simulation

The interactive simulator is below. Use the controls to explore the concepts described above.

 

Usage Instructions

1. Drag the fiber: Move the yellow control points to reshape the fiber. The core, cladding, and light paths follow the spline.
2. Change core and indices: Adjust n_core and n_clad. Larger index contrast increases numerical aperture and acceptance angle. The visible ray path updates immediately because the simulator uses both the internal refraction angle and the launch-to-acceptance margin.
3. Toggle clad presence: Turning Clad present off removes the physical guiding boundary in this model, so the light is shown escaping rather than undergoing total internal reflection.
4. Change launch angle: The colored arrow at the input shows the launch vector. If the launch angle exceeds the acceptance angle, the red light path escapes the fiber.
5. Use WDM: Set the center Wavelength, then increase Source to show nearby WDM channels propagating in the same fiber. Each color uses the same reflection rule.
6. Watch bend loss: Tight bends reduce the estimated output quality. The readout reports the minimum bend radius and bend/coupling loss.
7. Animate: Use Step Fwd, Step Bwd, or Run to move the optical pulses along the fiber.

Important Simplifications

This is an educational geometric-optics visualization, not a fiber-optics design tool:

• Ray optics, not wave optics. It does not solve Maxwell's equations or compute mode-field distributions; guided modes, cutoff conditions, and single-mode vs multimode behavior are stylized, not solved.
• No dispersion or polarization. Chromatic dispersion, polarization-mode dispersion, and material/waveguide dispersion are absent, so pulse spreading and WDM channel walk-off are not modeled.
• Approximate bend loss. Bend loss is a planning-style estimate from the minimum spline radius, not a wavelength-dependent macro/micro-bend loss from an electromagnetic mode solver.
• Idealized interfaces. Splice loss, connector reflection (return loss), and coupling efficiency are simplified or omitted.
• Exaggerated index range. The n_core/n_clad sliders use a wider range than real telecom fiber so the acceptance-cone effect is easy to see; values are illustrative, not standard SMF-28 numbers.
• Confinement heuristic. The displayed path adds a small educational confinement factor based on launch angle / theta_max rather than a physical power-overlap integral.