RF system for a 30 GHz, 5 TeV linear collider based on conventional technology

In order that it may be built within a reasonable length and with reasonable ac power consumption, a 5 TeV linear collider must employ an accelerating gradient and rf frequency which are both higher than for present 1 TeV collider designs. The required peak rf power per meter, which will also be higher than for 1 TeV designs, can be provided either by relatively conventional rf technology or by a two-beam scheme such as that proposed for CLIC. In this paper the first alternative, a 30 GHz rf system employing microwave tube power sources together with rf pulse compression, is described which produces an accelerating gradient on the order of 200 MV per meter. Limitations on the peak power that can be obtained from conventional klystrons as a function of frequency are discussed; it is found that such klystrons are only marginally adequate as a power source at 30 GHz. Several alternative rf sources, such as multiple-beam klystrons, sheet-beam klystrons, gyroklystrons and annular-beam ubitrons are described which are capable of providing the required power, after pulse compression, of about 600 MW per meter.