Fourier Series Builder
Learning goal
See Fourier series converge: each added harmonic sharpens the partial sum toward the target waveform, the spectrum bars show which coefficients do the work, and the stubborn overshoot at every jump — the Gibbs phenomenon — shows what convergence "almost everywhere" really means.
Things to try
- Square wave, N = 1 → 60. The wiggles crowd toward the jumps but the ~9% overshoot never shrinks. Pointwise vs uniform convergence, live.
- Triangle wave. Continuous function ⇒ coefficients fall as \(1/n^2\) ⇒ near-perfect fit by \(N = 10\). Compare the spectrum decay with the square wave's \(1/n\).
- Sawtooth — all harmonics present (odd and even). Which symmetry of the square/triangle waves kills their even terms?
- Rectangular pulse. The spectrum is a sampled sinc: narrow pulse ⇒ broad spectrum — the uncertainty principle in one picture.