Contents I. Introduction
Research into explosively driven magnetic flux compression generators [MFCGs, also referred to simply as flux compression generator (FCG)] has been ongoing since the early 1950s [1]. By the present time, the main operating principles and loss mechanisms of FCGs are reasonably well understood and documented in published works [1], [2], [3], [4], [5]. A number of magnetohydrodynamic (MHD) simulation codes have been developed over the years. All of them aim to accurately model the performance of FCGs. It was not until relatively recently, though, that generator performance was modeled with reasonable accuracy [6], [7]. This code, however, relied on the assumption that no internal breakdown occurs. The problem with this assumption is that in practice, shock loading of the gas ahead of the armature leads to a sharp rise in temperature and pressure and, subsequently, a high rate of ionization and an increase in conductivity [8], [9], [10]. This can trigger breakdown from the armature to the stator ahead of the armature contact point and significantly degrade generator performance. The exact thermodynamic state of the gas in this region is poorly documented in literature and is difficult to accurately model even in gases with a known equation of state (EOS). With the recent push to replace SF6 due to environmental concerns about its use and move toward new insulating gases such as Novec1, this becomes an even more difficult problem as there is not a known EOS on which to model the gas behavior. The sparse data already available indicate that Novec may perform superior to SF6 with respect to electrical breakdown [15]; however, these studies were conducted at low frequency (50–60 Hz) or dc.
Trademarked.
