Conceptual Designs for NLC Ubitrons with Permanent-Magnet Wigglers

This paper describes three embodiments of the ubitron (FEL) amplifier that will be analyzed for possible use on the NLC. The design frequency and power are 11.424 GHz and 200 MW peak rf output power. The baseline against which these conceptual designs are to be evaluated is the PPM-focused 50-MW SLAC klystron, which in simulation shows 65% efficiency. (See paper in this conference by Ken Eppley.) In order to remain competitive in cost and power consumption, only ubitron beam-wave configurations that can use permanent-magnet wigglers are considered. The plan for the NLCTA at SLAC is to use pulse compression to increase the 50-MW output from a klystron to 200-MW peak rf power with approximately 250-ns pulse length. If one had a 200-MW rf generator instead, how could it best be utilized? The 200-MW source could be substituted for the 50-MW klystron in one of two ways: 1. Eliminate the pulse compressor at some savings in both cost and system efficiency, but with no savings in the number of generators required for the accelerator. 2. Retain the pulse compressors but reduce the number of generators by a factor of four. Eliminating the pulse compressors allows one to accept a lower efficiency from the 200-MW generator without degrading the overall efficiency of the system, due to the elimination of the power loss in the pulse compressor. This is not, however, a pure gain. The 250-ns pulse would presumably retain the rise and fall time of the longer 1.5 μs pulses used with the 50-MW klystron, with the result that this wasted energy represents a higher percentage of the total energy in the pulse. To overcome this shortcoming implies the use of a gridded or mod anode gun which greatly complicates beam formation. Retaining the pulse compressors implies either the development of an 800-MW pulse compressor (four-to-one compression with 200-MW input) or the use of two or four outputs from the generator with a corresponding number of pulse compressors per tube. In this case, the remaining savings is that associated with the reduction in the number of generators required, a reduction which might indeed be significant.