Solution of Maxwell's equations: some practical problems in applying present-day codes

The main purpose of the article is not to discuss the solution of Maxwell's equations or of EMC problems in any sort of mathematical depth, but to point out some of the pitfalls which await the unwary when they apply present-day EMC modelling codes to the solution of problems with a real-life degree of complexity. The intention is not to suggest that such codes are not valuable; but rather to point out the crucial need for experience, and the relative merits of computation and testing. Most of the pitfalls discussed have their origins either in the complex geometry of real electronic systems or in the difficulty/expense of characterising parts of those systems. We begin with an overview of the general classes of EMC modelling codes; without making any real distinction between those codes which are still under development (called laboratory codes) and commercially available codes.