AbstractElectromagnetic compatibility has been of crucial importance to fusion experiments, long before ot was known by its present name, EMC.The huge amounts of pulsed power required to create and heat a plasma, and the stray fields produced in large volumes made it difficult, in earlier and in present experiments, to reach the ideal EMC-state where interference does not hinder (or impede) operation, measurements and data acquisition. The pressure to obtain results from the experiments has led to many ingenious solutions, but left usually little time for a more fundamental study.A basic problem is that the complex three-dimensional (and often toroidal) circuitry of a fusion experiment cannot be correctly represented by network theory models. In particular extended leads no longer “transport” potentials or voltages, but of course continue to transport currents. We therefore concentrate on currents and on all closed current loops rather than on worrisome (but meaningless) “jumping potentials”.With concepts developed along these lines we can design grounding systems which protect large interconnected electrical systems against interference. Such a grounding system may use wide strips, conduits or tubes to carry all interconnecting cables from subsystem to subsystem. Alternative descriptions of this grounding system may help to clarify its operation: a screen room reduced to its essential current-carrying elements, a minimization of transfer impedances or a structure to keep the common mode circuits closely under control. In this context ground loops can be quite useful, and we abandon the concept of single-point grounding.Next to the protection by grounding and shielding, additional measures can be taken to carry signals safely into subsystems. Larger amplitude or differentiated signals are advantageous, provided the first attenuator or integrator section in the “receiver” uses only passive, linear components.EMC is essential for fusion experiments, as stated in the title. Techniques borrowed from fusion can however also be quite useful in other fields. Examples in power engineering measurements and in lightning studies are briefly mentioned.
Diagnostics: Chapter 8 of the special issue: on the path to tokamak burning plasma operation