Source

Adapted from my own handwritten and hand-typed notes from Year 1 to Year 3 of my EEE degree

Separation of variables assumes the solution is a product of functions of one variable each, which turns one PDE into separate ODEs.

This is a strong assumption, so it only finds some solutions, and it only works for a narrow class of problems.

When it works

Conditions:

  • the PDE must be linear and homogeneous (we later add solutions by superposition)
  • the boundary conditions must be homogeneous (equal to 0), so they can be moved onto alone: with gives

Three steps

Steps:

  1. Substitute , divide to put all on one side and all on the other, and set both sides equal to a separation constant . This gives two ODEs
  2. Turn the boundary conditions into conditions on and solve the problem. Only some values of give nonzero
  3. Solve the equation, form the series of all product solutions, and fit the initial condition

The key reasoning in step 1: a function of only can equal a function of only for all and only if both are the same constant.

ODE revision needed

ODEGeneral solution

Worked example (first order, general functions)

Solve .

Solution:

  • let :
  • separate:
  • gives , and gives
  • , with and constants

The full three step method with boundary conditions is shown in Heat Equation by Separation of Variables.