Jay B. Chase

An Armature-Stator Contact Resistance Model for Explosively Driven Helical Magnetic Flux Compression Generators

Although helical magnetic flux compression generators (HFCGs) have been in use for more than four decades, no one has been able to satisfactorily model their behavior. To bring computed currents into agreement with experimental values, tuning factors or anomalous flux loss factors are used. Such factors are not universal, and they must be adjusted for each generator design, or for different operational parameters (e.g., seed current or load inductance) for a given design. Many HFCG modeling codes have been reported over the years with various types of these empirical factors. One of the recognized issues for HFCGs is magnetic flux loss near the moving contact point between expanding armature and helical stator coil winding. In our new model, we have analytically estimated the rate of magnetic field diffusion in the vicinity of the contact point. When converted to a flux loss rate, we find that it usually scales nonlinearly with the instantaneous current, and that the resulting effective resistance is proportional to the square root of the current. This result applies even at relatively small operating currents. Whereas the usual HFCG resistances drop as the generator length decreases, the contact resistance generally increases throughout operation. While small initially, we find that it usually dominates late in time and ultimately limits the gain of most generators. In this paper, we present the derivation of the contact resistance model and show its effectiveness in estimating current gain for simple HFCG designs using a simple spreadsheet program. The model has also been implemented in the 11/2-D FCG-model code, CAGEN, and an accompanying paper presents CAGEN results for a wide range of HFCGs, benchmarking the new model. The formulation for our model is universal; i.e., there are no adjustable factors, and it has generally enabled calculation of HFCG currents to within 20% of experimentally reported values.

Benchmarking the Care'n Co. Flux Compression Generator Code, CAGEN, Implementing the Kiuttu Contact Resistance Model

The PC-based program CAGEN has been described before and has since continued development. The most recent innovation implemented in CAGEN is the Kiuttu Contact Resistance Model (KCRM). This model is described elsewhere in these proceedings and represents the most important step forward for FCG modeling observed in many years. In this paper, the performance of a very large range of helical flux compression generators is computed using CAGEN, with no adjustable tuning factors. These generators span from the small Lawrence Livermore National Laboratory Minigen and the Lobo, each less than 100 cubic centimeters volume, to the quite large Los Alamos National Laboratory Mark IX, which is more than 220 liters in size. The examples range over a factor of 1000 in output current and over a factor of 100,000 in output energy, and represent different construction techniques. The results of eight such benchmark calculations using CAGEN, with the KCRM, are never in error more than 18% with respect to reported experimental current values.

The precision fabrication of non-split-turn, bare wire, constant pitch high explosive flux compression generators for use in internal electric breakdown experiments

In order to have sufficient precision to make experimental studies of the internal electric breakdown in flux compression generators, units were constructed using CNC techniques. The requirements were to have bare copper wires for the stator, aluminum armatures and transparent support materials. Reproducibility was critical. The details will be discussed along with metrology of the resulting devices.

Experiments with explosive flux compression Generators Confirming the CAGEN-CALE threshold predictions for internal electric breakdown

A total of seven shots were fired at the HyperSpectral Sciences, Inc. firing site in Cinebar, WA, in support of the program to elucidate the internal electric breakdown phenomena. Three experiments were full-up HFCG shots. All of the HFCG shots broke down, while the last one exhibited the clearest example of “classical” breakdown. The correlation of the computer model threshold calculation with the last shot is quite good. The use of framing and streak cameras as a diagnostic has proved viable; the contact point as well as the breakdown are very visible. It was found that the electrical arcs that result from the breakdowns do not move. They are overrun by the contact point. The compressed gas in the interior of the HFCG must be taken into account. While it seems quite evident that the breaks are from the stator to the armature, our evidence cannot prove the point. Details of the firing arrangements, the data recovery methods, and the results with comparisons to theory [1], [2], and [3] are presented.