Fast 704 MHz Ferroelectric Tuner for Superconducting Cavities
The Omega-P SBIR project described in this Report has as its goal the development, test, and evaluation of a fast electrically-controlled L-band tuner for BNL Energy Recovery Linac (ERL) in the Electron Ion Collider (EIC) upgrade of the Relativistic Heavy Ion Collider (RHIC) at Brookhaven National Laboratory (BNL). The tuner, that employs an electrically-controlled ferroelectric component, is to allow fast compensation to cavity resonance changes. In ERLs, there are several factors which significantly affect the amount of power required from the wall-plug to provide the RF-power level necessary for the operation. When beam loading is small, the power requirements are determined by (i) ohmic losses in cavity walls, (ii) fluctuations in amplitude and/or phase for beam currents, and (iii) microphonics. These factors typically require a substantial change in the coupling between the cavity and the feeding line, which results in an intentional broadening of the cavity bandwidth, which in turn demands a significant amount of additional RF power. If beam loading is not small, there is a variety of beam-drive phase instabilities to be managed, and microphonics will still remain an issue, so there remain requirements for additional power. Moreover ERL performance is sensitive to changes in beam arrival time, since any such change is equivalent to phase instability with its vigorous demands for additional power. In this Report, we describe the new modular coaxial tuner, with specifications suitable for the 704 MHz ERL application. The device would allow changing the RF-coupling during the cavity filling process in order to effect significant RF power savings, and also will provide rapid compensation for beam imbalance and allow for fast stabilization against phase fluctuations caused by microphonics, beam-driven instabilities, etc. The tuner is predicted to allow a reduction of about ten times in the required power from the RF source, as compared to a compensation system with narrower bandwidth. It is planned to build a 704 MHz version of the tuner, to check its underlying principles, and to make high-power tests at power densities aimed towards controlling 50 kW of average power. Steps towards this goal will be limited by, among other factors, losses in the actual ferroelectric elements in the ferroelectric assemblies. As the ferroelectric material loss tangent is reduced through efforts by the supplier Euclid TechLabs LLC, the concomitant power loss in its ferroelectric assemblies will drop, and the average power-handling capability of the Omega-P tuner will rise. It can thus be anticipated that the Phase II development project of the 704 MHz tuner will be iterative, but the pace and ultimate power-handling level of the tuner is difficult to predict at this early stage in Euclid's development program. Fortunately, since Omega-P's conceptual tuner is a simple module (nominally rated for 5 kW), so that the number of modules required in each tuner can be chosen, depending upon the cavity power level needed, plus the power for tuner losses.