Relations between the major size and cost parameters of tokamaks

A set of preliminary estimates of the effects of toras size on major design and cost parameters was made for a series of tokamak reactors with superconducting coils for the toroidal field. The three sets considered include essentially zero power scientific feasibtlity experiments, low-power reactor systems for plasma physics experiments, and full-scale power reactors. The prime concern is with the cost of the superconducting magnet coils, but allowances for the blanket and magnet shield systems are also included as well as the system for heating the plasma to the ignition temperature. The prime objective was to show trends for major parameters, not absolute values for costs. The results of the study indicate that the cost of a superconducting coil system for the toroidal field increases relatively little with torus diameter for a scientific feasibility experiment, is quite insensitive to torus size for a plasma physics test reactor, and actually falls off with increasing size for a full scale power reactor. These effects stem from the reduction in the magnetic field required associated with an increase in torus size together with the increase in the allowable current density in the superconductor material with a reduction in the design value for the magnetic field. The results indicate further that, for a modest increase in the cost of superconducting magnets for the toroidal field, it should be possible to build a feasibility experiment so that it can be converted readily into a plasma physics test reactor. (auth)