Choice of Geometry for the Alvarez Section of the FMIT Deuteron Linac

Abstract
The Hanford Fusion Materials Irradiation Test (FMIT) 35-MeV linac is intended to produce neutrons for materials studies. Its operation will be virtually continuous for at least 20 years. Such operation implies that the accelerator design must be conservative to avoid excessive downtime and that the accelerator be economical of power. The initial design of the linac drift-tube section was preceded by more than 800 SUPERFISH computer runs. SUPERFISH allows the selection of a geometry for the FMIT machine that has very nearly the maximum value of ZT2 consistent with Kilpatrick's criterion. This maximum value insures that power costs will be as low as possible while keeping the maximum surface fields at a level conservative enough to prevent sparking. The optimization procedure is examined to show how the best geometry is achieved. The drift-tube section will operate at 80 MHz with an average 1.4 MV/m gradient. As presently conceived, it will consist of two tanks, one 248 cm in diameter and about 18.-m long and the other 240 cm in diameter and about 14.-m long. The number of drift-tubes will be 73 if a stable phase of -30° is chosen.

This publication has 2 references indexed in Scilit: