We present experiments and simulations showing the behavior of a free-electron laser ͑FEL͒ with both positive and negative linear tapers along the wiggler. We show the power desynchronism curve widths, efficiency, exhaust electron energy spread, and wavelength dependence as a function of taper for 3-and 6-m optical wavelengths and for resonators with 10% and 2% loss/pass. Simulations of the experiments, using a multimode analysis, are seen to be in general agreement with the experimental results, carried out at the IR Demo FEL at Thomas Jefferson National Accelerator Facility. We find that short-pulse effects are more effective than tapers in producing high efficiency with low exhaust energy spread, and the expected performance enhancement of FEL tapering is not achieved.

The TJNAF Free Electron Laser (FEL) will be upgraded to operate at 10 kW average power in the near future.Multimode simulations are used to analyze the operation describing the evolution of short optical pulses in the far infrared wavelength regime. In an FEL that recirculates the electron beam, performance can depend on the electron beam distribution exiting the undulator. The effects of varying the undulator field strength and Rayleigh length of the resonator are explored, as well as the possibility of using an optical klystron. The simulations indicate that the FEL output power can reach the design goal of 10 kW. r

Experiments using the TJNAF FEL have explored the operation with both tapered and inversely tapered undulators. We present here numerical simulations using the TJNAF experimental parameters, including the effects of taper. Singlemode simulations show the effect of taper on gain. Multimode simulations describe the evolution of short optical pulses in the far infrared, and show how taper affects single-pass gain and steady-state power as a function of desynchronism. A short optical pulse presents an ever-changing field strength to each section of the electron pulse so that idealized operation is not possible. Yet, advantages for the recirculation of the electron beam can be explored. r

Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instruction, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to