Speaker
Description
Cavity desynchronization is important in FEL oscillators, especially in the IR and THz regimes where optical slippage is significant and laser lethargy can reduce gain under perfect synchronization. Because fast mechanical cavity tuning is difficult, several facilities have implemented dynamic desynchronization by sweeping the RF frequency of the accelerator, typically using a linear frequency ramp across a macropulse. Using the University of Hawai‘i at Mānoa FEL oscillator as an example system, we treat desynchronization as a pass-dependent control variable and compare optimized linear ramps with gain-informed profiles and adaptive optimization control in full 3D time-dependent simulations. The dynamic strategies substantially increase the accumulated pulse energy over the macropulse relative to perfect synchronization and outperform the static cases considered here, while the different dynamic approaches themselves give comparable performance. The dynamic strategies also recover the characteristic temporal and spectral spiking structure of the moderately desynchronized saturated FEL oscillator. These results show that dynamic desynchronization can substantially improve FEL oscillator performance, while a predetermined linear ramp can capture most of the benefit without the added experimental and computational complexities of the gain-informed and the adaptive optimization profiles.
Funding Agency
Supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences and Office of Accelerator Research & Development and Production, under Contract No. DE-SC0025583.
| I have read and accept the Privacy Policy Statement | Yes |
|---|