Abstract
Event location has been used in the implementation of an existing finite-difference method for phase-change problems. The finite-difference method results in a model that changes structure every time the interface crosses a spatial grid line. In the traditional methods, either a fixed time step has been used, or the time step has been calculated by an iterative procedure as the time for the interface to move a single space increment. By using event location, the implementation stage is simplified, and the size of the time steps is automatically adapted to the interface dynamics.