Some of the computational aspects which may affect the validity and applicability of a numerical solution for a th1n-w1re structure are considered 1n this paper. These Include: (1) structure segmentation, (2) current expansions, (3) the th1n-w1re approximation, (4) matrix factorization roundoff error, (5) near-field numerical anomalies, (6) multiple junction treatment, (7) w1re-gr1d modeling, and (8)

A two-year study of recirculating induction heavy ion accelerators as low-cost driver for inertial-fusion energy applications was recently completed. The projected cost of a 4 MJ accelerator was estimated to be about $500 M (million) and the efficiency was estimated to be 35%. The principal technology issues include energy recovery of the ramped dipole magnets, which is achieved through use of ringing inductive/capacitive circuits, and high repetition rates of the induction cell pulsers, which is accomplished through arrays of field effect transistor (FET) switches. Principal physics issues identified include minimization of particle loss from interactions with the background gas, and more demanding emittance growth and centroid control requirements associated with the propagation of space-charge-dominated beams around bends and over large path lengths. In addition, instabilities such as the longitudinal resistive instability, beam-breakup instability and betatron-orbit instability were found to be controllable with careful design.

We report on progress in the preparation of the 2 MV Injector at LBNL as the front-end of Elise, and as a multi-purpose experimental facility for Heavy Ion Fusion beam dynamics studies.Recent advances on the performance and understanding of the injector are described, and some of the on-going experimental activities are summarized. 1.

Some of the computational aspects which may affect the validity and applicability of a numerical solution for a th1n-w1re structure are considered 1n this paper. These Include: (1) structure segmentation, (2) current expansions, (3) the th1n-w1re approximation, (4) matrix factorization roundoff error, (5) near-field numerical anomalies, (6) multiple junction treatment, (7) w1re-gr1d modeling, and (8)

Tlfc is i fKfmt .f 1 pftr mtnki br (ttH-• itiijl-••? te wfc tafa. HBoli-. •• Htpmt «•*-**«» *e-" nil m W art » wfiifctrf lifcir *i friM»in «r*ci Knnmi «F ns rawai s unfa ^