Further Development, Operation and Experiments with the IRD Closed-Cycle MPD Loop

The IRD closed-cycle MPD loop has been operated as a power generator since April 1964 and several hundred separate experiments have been carried out. Throughout this period loop development has been continued to establish a system to give accurate and reproducible electrical and physical data. Apart from the MPD generation experiments, the IRD facility is unique in that many aspects of high- temperature loop technology have been explored. Much of the information derived from this programme is relevant to the construction of large-scale systems. The loop uses helium up to temperatures of 2200°C . with controlled caesium seeding. The mixture enters the duct through a subsonic nozzle. The 22 kG magnet has an axial field length of 22 in. After the diffuser recuperative heat exchanger and cooler, caesium is removed in traps and filters before the helium re-enters the circulator. Transport and reaction . mechanisms within the high-temperature regions resulting in formation and deposition of electrically- conducting layers on the generator duct walls have been investigated. The present loop has operated successfully and remained clean after frequent cycles and periods of high-temperature operation exceeding 50 hours. Generator duct wall heaters give control of the wall temperature; in particular, isothermal . conditions are obtainable with more precise definition of boundary-layer and plasma temperature parameters. High-temperature windows have allowed pyrometric temperature measurement of nozzle inlet stagnation temperature, visual and spectroscopic viewing of the caesium vapour injection and helium-caesium plasma, and microwave and spectroscopic analysis of plasma parameters. Investigations have been carried out on problems of vaporizing and superheating caesium seed in the high-temperature helium flow, caesium seed recovery, filtration and purification. Improved helium purification and analysis plant incorporating a fully automatic chromatograph sampling valve provides impurity gas analysis to levels of less than 1 ppm. Such information is essential in assessing the effects of molecular impurities on electron energy loss and non-equilibrium ionization. The variation of electrical conductivity has been investigated: the parameters include stagnation gas temperature, stagnation gas pressure, channel Mach number, magnetic field strength, electrode to insulator width ratio, electrode width to channel diameter ratio, channel wall temperature, helium mass flow, caesium seeding fraction and helium purity level. All experiments have been performed with constant cross-sectional area channels, both rectangular and circular, and with alumina and boron nitride insulator walls, and tantalum and stainless steel electrodes. Segmented Faraday- and Hall-modes of operation have been investigated. For most operating conditions with the Faraday mode electrical conductivities corresponding to thermal equilibrium ionization have been observed; under certain conditions of electrode segmentation and helium purity, significant magnetically-induced extrathermal ionization does occur. This is discussed. Only limited success in the Hall mode has been achieved. In the majority of the experiments the measured open-ciicuit voltage is considerably less than anticipated; the reasons for this discrepancy are examined in detail. In certain respects the segmented Faraday mode of operation approximates that of a solid electrode Faraday generator. For example, when thermal ionization occurs the electrical conductivity, in several cases decreased with increasing magnetic field. In all cases the generated current density is sufficiently small to be provided by thermionic emission from the electrodes if total or partial caesium coverage occurs. No evidence for electrode (or insulator) degradation has been found even after extended periods of operation. (author)